cuii and znii coordination chemistry of pyrazole-containing polyamine receptors − influence of the...

FULL PAPER

CuII and ZnII Coordination Chemistry of Pyrazole-Containing PolyamineReceptors � Influence of the Hydrocarbon Side Chain Length on the Metal

Coordination

Carlos Miranda,[a] Francisco Escartı,[b] Laurent Lamarque,[a] Enrique Garcıa-Espana,*[b]

Pilar Navarro,*[a] Julio Latorre,[b] Francisco Lloret,[b] Hermas R. Jimenez,[b] andMarıa J. R. Yunta[c]

Dedicated to Prof. Dr. Jose Elguero

Keywords: Pyrazole macrocycles / Copper and zinc coordination / Acid-base and metal formation equilibria / Polyamineligands / Paramagnetic NMR

The synthesis of a new macrocyclic receptor (L4) containingtwo 3,5-dimethylpyrazole units connected by dipropylenetri-amine bridges is reported for the first time; pH-metric titra-tions indicate that L4 shows six protonation steps in the pHrange 2−11. In the absence of metal ions, the pyrazole moiet-ies are not involved in acid-base processes in this pH range.Addition of CuII and ZnII results in deprotonation of the pyra-zole moieties which act as bis(monodentate) η1:η1 ligands.This induced deprotonation occurs at higher pH values thanin the complexes of the analogous ligand containing diethyl-enetriamine bridges (L1). The crystal structures of[Cu2(H−2L4)](ClO4)2 and [Zn2(H−2L4)](ClO4)2 obtained bytreatment of Na2[H−2L4] with either Cu(ClO4)2·6H2O orZn(ClO4)2·6H2O in methanol followed by recrystallisationfrom water/methanol also show the deprotonation of the pyr-azole moieties. The coordination geometry around eachmetal ion is square-pyramidal and involves all nitrogenatoms of the macrocycle. The crystal structure of[Zn2(H−2L1)](ClO4)2 shows full involvement of all the nitro-gen atoms of the macrocycle in the coordination to the metalions. The coordination geometry can be defined as midwaybetween a square pyramid and an irregular trigonal bipyr-amid. Treatment of neutral L4 in methanol withCu(ClO4)2·6H2O yields a blue complex which, after heating

[a] Instituto de Quımica Medica, Centro de Quımica OrganicaManuel Lora Tamayo, CSIC,c/ Juan de la Cierva 3, 28006 Madrid, SpainE-mail: [email protected]

[b] Departamento de Quımica Inorganica, Facultad de Quımica,Universidad de Valencia,c/ Doctor Moliner 50, 46100 Burjassot (Valencia), SpainFax: (internat.) � 34-96-3864322E-mail: [email protected]

[c] Departamento de Quımica Organica, Facultad de Quımica,Universidad Complutense de Madrid,Avda. Complutense s/n, 28040 Madrid, SpainSupporting information for this article is available on theWWW under http://www.eurjic.org or from the author.

Eur. J. Inorg. Chem. 2005, 189�208 DOI: 10.1002/ejic.200400671 © 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim 189

in boiling water, crystallises as a red compound. Elementalmicroanalyses, ESI-MS and FAB-MS data of both forms ofthis complex are consistent with the formula[Cu2L4](ClO4)4·2H2O. Furthermore, these data along with theparamagnetic 1H NMR spectrum of the red form of the com-pound suggest a structure in which the pyrazole rings aredeprotonated while the central nitrogen atoms of thebridging chains are protonated and consequently uncoordi-nated. The change from the blue (square-pyramidal) to the redform (square-planar) can be ascribed to dissociation of thewater molecules which occupy the axial positions in the CuII

coordination spheres of the blue form. The variation of themagnetic susceptibility of the red complex [Cu2L4]-(ClO4)4·2H2O and the complex [Cu2(H−2L4)](ClO4)2 with tem-perature has been studied in the 2−300 K temperature range.Fitting of the experimental data provides J values whichare among the highest found for doubly pyrazolate-bridgeddicopper(II) complexes {J = −299 cm−1 for red-[Cu2L4](ClO4)4·2H2O and J = −286 cm−1 for [Cu2(H−2L4)]-(ClO4)2}. A trinuclear CuII complex of formula [Cu3-(H−2L4)](ClO4)4·2MeOH was also isolated after treatment ofL4 with Cu(ClO4)2·6H2O solutions of higher concentration.(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,Germany, 2005)

IntroductionOver the last decades a large amount of research effort

has been devoted to identifying the biological roles of metalions such as CuII and ZnII. In order to achieve such a goal,different parallel approaches have been followed. The moststraightforward consists of directly studying the active sitein the metalloenzymes. A second approach, which has pro-ved very useful in some instances,[1�3] has been to mimicthe chemical features of the metal sites using smaller sizedcoordination compounds as models. The knowledge ob-

E. Garcıa-Espana, P. Navarro et al.FULL PAPERtained from the way in which metalloenzymes activate smallmolecules like H2, N2, CO2, etc. has also resulted in con-siderable interest in the development of catalysts for indus-trial purposes.

A second biological aspect of metal complexes that hasto be taken into account is their implications in pharma-cology. For instance, complexes of macrocyclic ligands suchas bicyclams and their ZnII complexes have shown usefulapplications as potent anti-HIV-1 agents and several MnII

and CuII complexes of hexaazamacrocycles have found ap-plications as SOD mimics.[4] Another point of interest con-cerns the use of chelate ligands for selectively complexingand removing excess amounts of metal ions from whicharise undesirable effects such as the oxidative damage de-tected in several neurodegenerative disorders.[5] An illustra-tive example comes from the use of the ligand PBT1 (cli-oquinol) for removing excess CuII which accumulates in theneocortex following the oxidative damage produced inAlzheimer disease.[6]

Therefore, the preparation and study of the coordinationchemistry of ligands featuring new characteristics in theirinteraction with metal ions represents a challenge in coordi-nation and biomedical chemistry. Ligands including 1H-pyrazole fragments offer a wealth of possibilities in this re-spect. The pyrazole unit can behave as a monodentateligand through its sp2 nitrogen atom or, upon depro-tonation, as an η1:η1 bis(monodentate) bridging ligand(Scheme 1).[7,8]

Recently, we have examined the acid-base behaviour, CuII

coordination chemistry and dopamine recognition capabili-ties of a series of pyrazole-containing polyamine macro-cycles of different topologies [see ligands 1 (L1) to 3 (L3) inScheme 2].[9�11] Here we extend these studies to the metalion ZnII. Additionally, we report on the synthesis of thepyrazole-containing macrocycle 3,7,11,14,15,18,22,26,29,30-decaazatricyclo[26.2.1.113,16]dotriacontane-1(31),13,

Scheme 2

© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.eurjic.org Eur. J. Inorg. Chem. 2005, 189�208190

Scheme 1

16(32),28-tetraene (4) (L4) (see Scheme 2) which has onlypropylenic chains and the preparation of its disodium dipyr-azolate salt 4� [Na2(H�2L4)]. We discuss the effect of thepropylenic chains on the acid-base behaviour and on theCuII and ZnII coordination chemistry in aqueous solution.We also report different CuII and ZnII complexes preparedeither from 4 and 4� or from the ligands 1�3 and we discussthe influence of the hydrocarbon chain length on the coor-dination features of these new complexes.

Results and Discussion

Synthesis of 4 and 4�

The preparation of 4 was performed by a [2�2] dipodalcondensation of 3,5-pyrazoledicarbaldehyde with N-(3-ami-nopropyl)-1,3-propanediamine in methanol at room tem-perature, followed by in situ reduction of the resultantSchiff base with NaBH4. After careful purification firstly bychromatography and then by crystallisation from toluene, 4(L4) was isolated as a pure solid of melting point 154�156°C in 49% yield. Compound 4 and its hydrochloride 4·6HCl

CuII and ZnII Coordination Chemistry of Pyrazole-Containing Polyamine Receptors FULL PAPER

Table 1. Comparison of 1H and 13C NMR spectroscopic data (δ [ppm], D2O) of 4 recorded at pH � 10.5 and 8.0 and 4·6HCl recordedat pH � 3.7

4 (1) pH � 10.5 4 (2) pH � 8.0 4·6HCl (3) pH � 3.7 ∆δ (2�1) ∆δ (3�2)

4-H 6.07 (s, 2 H) 6.22 (s, 2 H) 6.63 (s, 2 H) 0.15 0.416-H 3.53 (s, 8 H) 3.71 (s, 8 H) 4.26 (s, 8 H) 0.18 0.558-H 2.26 (t. 8 H)[a] 2.61 (t, 8 H)[a] 3.05 (m, 8 H) 0.35 0.449-H 1.37 (m, 8 H) 1.70 (m, 8 H) 2.01 (m, 8 H) 0.33 0.3110-H 2.26 (t. 8 H)[a] 2.87 (t, 8 H)[a] 3.05 (m, 8 H) 0.61 0.18

4 (1) pH � 10.5 4 (2) pH � 8.0 4·6HCl (3) pH � 3.7 ∆δ (2�1) ∆δ (3�2)

C-3,5 147.71 145.40[b] 139.20 �2.31 �6.20C-4 103.68 104.52 109.01 0.84 4.58C-6 43.79 43.64 42.68 �0.15 �0.96C-8 45.34 45.26 43.83 �0.08 �1.43C-9 28.40 24.79 22.81 �3.61 �1.98C-10 46.55 45.96 44.65 �0.59 �1.31

[a] J � 7 Hz. [b] Broad signal.

(m.p. 265�268 °C) have been fully characterised by elemen-tal analyses and NMR spectroscopy.

Table 1 shows a comparison of the 1H and 13C NMRspectroscopic data of the new ligand 4, recorded in D2O atpH � 10.5 and 8.0, with those of its hexahydrochloride4·6HCl recorded at pH � 3.7. The signals have been ana-lysed on the basis of their HMQC and multiple-bondHMBC 2D NMR heteronuclear correlation spectra. Theabove 1H NMR spectra are graphically compared in theSupporting Information [Figure 1S (A�C)] (for SupportingInformation see also the footnote on the first page of thisarticle).

As usually occurs for 3,5-disubstituted 1H-pyrazole de-rivatives, at pH � 10.5 (where the nonprotonated speciespredominates) 4 shows a very simple pattern of signalswhich indicates that because of the fast prototropic equilib-rium of the pyrazole ring, the compound exhibits an aver-age fourfold symmetry on the NMR time scale. The carbonatoms labelled C-3 and C-5 (for the labelling see Scheme 2)appear together as a sharp signal and the pairs of methylenecarbon atoms C-6, C-8, C-9 and C-10 are magneticallyequivalent.

In general, the pH-dependent variation of the NMR res-onances of the β-carbon atoms and of the hydrogen atomsbound to the α-carbon atoms with respect to the protonatednitrogen atoms are of great help in establishing protonationsequences of polyamine molecules.[9] As can be derivedfrom the basicity constants, the two first protonations of L4

occur in the pH range of 11�8. Thus, in the NMR spectraof 4 recorded at pH � 8.0 (Table 1), the carbon signalshifted the most upfield is that of carbon atoms C-9 (∆δ ��3.61 ppm), whereas the signal of carbon atoms C-8 andthat of the quaternary carbon atoms C-3,5 do not experi-ence significant shifts in this pH region. In the same way,the most downfield shifted 1H signal is that of 10-H (∆δ �0.61 ppm), clearly indicating that 4 is protonated firstly atthe centre nitrogen atoms of the bridging chains. On theother hand, by comparing the NMR signals of 4 at pH �8 and those of 4·6HCl at pH � 3.7 (Table 1) it can be ob-

Eur. J. Inorg. Chem. 2005, 189�208 www.eurjic.org © 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim 191

served that after protonation of the aliphatic nitrogen atomscloser to the pyrazole rings, the most downfield shifted hy-drogen signals are those of 6-H and 8-H (∆δ � 0.55 and0.44 ppm, respectively) and that the signals of the methyl-ene carbon atoms which experience the largest upfieldvariations are C-9 � �1.98 ppm) which are located in theβ-position with respect to the above-mentioned nitrogenatoms.

Furthermore, in 4·6HCl, the signals of the pyrazole car-bon atoms C-3,5 and C-4 (located in the β- and γ-positions,respectively, in relation to the closer protonated aliphaticnitrogen atoms) also experience large upfield (∆δ �6.20 ppm for C-3,5) and downfield (∆δ � 4.58 ppm for C-4) shifts. These variations are of a similar order to thosepreviously reported for L1.[9] It is important to point outthat such variations do not reflect any particular impli-cation of the pyrazole subunits in the protonation process,since it is well known that β- and γ-aromatic carbon atomsbelonging to polyamine-bridged cyclophanes containingbenzene units also bear similar chemical shifts upon pro-tonation.[12,13] Finally, we also obtained the disodium dipy-razolate salt 4� [Na2(H�2L4)] as a pure solid (m.p. 208�210°C) in 98% yield, by heating the free ligand 4 to reflux with2 equiv. of NaOH in MeOH.

Acid-Base Behaviour

Table 2 includes the stepwise protonation constants ofthe new receptor L4 and those previously reported for com-pounds 1 (L1) to 3 (L3).[9�11] The basicity constants of 4(L4), with only propylenic chains are higher than those ofL1 containing only ethylenic chains and are also higher thanthose of the compound with ethylenic chains and N-benzyl-pyrazole spacers (L3). The differences between the stepwiseconstants of L4 and those of L1 and L3 are particularlylarge for the fifth and sixth protonation steps (Table 2). Thefifth and six protonations occur at the central nitrogenatoms of the polyamine chains of the macrocycle. Therefore,at this stage, the incoming proton will notice a lower elec-

E. Garcıa-Espana, P. Navarro et al.FULL PAPER

Table 2. Protonation constants of the receptors L1 to L4 determinedin 0.15 mol·dm�3 NaCl at 298.1 K

Reaction[a] L1 [b] L2 [c] L3 [b] L4 [d]

H � L � HL 9.74 9.54 8.90 10.05(2)[e]

H � HL � H2L 8.86 8.71 8.27 9.34 (2)H � H2L � H3L 7.96 7.72 6.62 7.63 (3)H � H3L � H4L 6.83 6.55 5.85 7.07 (3)H � H4L � H5L 4.57 6.52 3.37 6.45 (3)H � H5L � H6L 3.19 5.22 2.27 5.55 (4)log β[f] 41.1 44.3 35.3 46.1

[a] Charges omitted for clarity. [b] Taken from ref.[9] [c] Taken fromref.[11] [d] This work. [e] Values in parentheses are standard devi-ations in the last significant figure. [f] log β � Σlog KHjL.

trostatic repulsion from the adjacent polyammonium sitesin L4 than in L1 or in L3 due to the presence of the largerpropylenic chains.[14] Curiously, the basicities displayed byL4 and the cryptand L2 are rather similar even if their struc-tures and numbers of nitrogen atoms are different.

The high degree of protonation that L4 presents atphysiological pH values (mean number of protons at pH �7.4 is 3.02) makes this receptor a promising candidate forthe coordination of anionic or polar species.

Metal Coordination in Aqueous Solution

The stability constants for the formation of CuII com-plexes of 4 (L4) together with those previously reported forthe related receptors 1 (L1) to 3 (L3) under the same exper-imental conditions are shown in Table 3.[11] The stabilityconstants for the interaction of ZnII with L1 to L4 deter-mined in this work are presented in Table 4.

Table 3. Stability constants for the formation of CuII complexes of the receptors L1, L2, L3 and L4 determined in 0.15 mol·dm�3 NaClat 298.1 K

Entry Reaction[a] L1 [b] L2 [b] L3 [b] L4 [c]

1 Cu � L � 3 H � CuH3L 37.68(9)2 Cu � L � 2 H � CuH2L 35.05(2)[d] 35.42(8) 27.9(1) 32.64(6)3 Cu � L � H � CuHL 30.82(2) 28.85(7) 22.12(8) 25.91(4)4 Cu � L � CuL 23.48(3) 21.76(6) 13.60(7) 17.83(8)5 Cu � L � CuH�1L � H 13.05(3)6 2 Cu � L � H � Cu2HL 39.04(7) 29.37(5)7 2 Cu � L � Cu2L 34.45(3) 36.72(3) 26.27(3) 28.72(2)8 2 Cu � L � Cu2H�1L � H 31.31(5) 16.85(4) 23.62(3)9 2 Cu � L � Cu2H�2L � 2 H 26.56(3) 24.13(5) 7.06(3) 15.69(8)10 2 Cu � L � Cu2H�3L � 3 H 5.87(7)11 CuH2L � H � CuH3L 5.0412 CuHL � H � CuH2L 4.23 6.57 5.78 6.7313 CuL � H � CuHL 7.09 8.52 8.0814 CuL � CuH�1L � H 7.34 8.7115 CuL�Cu � Cu2L 10.97 12.67 10.8916 Cu2L � H � Cu2HL 2.32 3.117 Cu2L � Cu2H�1L � H �5.41 �9.42 �5.118 Cu2H�1L � Cu2H�2L � H �7.18 �9.79 �8.0719 Cu2L � Cu2H�2L � 2 H �7.89 �12.59 �19.21 �13.0320 Cu2H�2L � Cu2H�3L � H �9.72

[a] Charges omitted for clarity. [b] Taken from ref.[11] [c] This work. [d] Values in parentheses are standard deviations in the last signifi-cant figure.


With CuII, L4 forms mononuclear [Cu(HxL)](x�2) com-plexes with x � 0, 1, 2 and 3 and dinuclear [Cu2(HxL)](x�4)

species with x � �1, �2 and �3. The extent of formationof each one of these complexes depends on pH, M/L molarratio and concentration as can be seen in Figure 1 whichshows the distribution diagrams for the CuII complexes ofL4 in molar ratios of 1:1 (Figure 1, A) and 2:1 (Figure 1,B), complexes of L1 in a molar ratio of 2:1 (Figure 1, C)and for the ZnII complexes of L4 and L1 in a molar ratio2:1 (Figure 1, D and E, respectively).

The stability constant for the species [CuL4]2� is inter-mediate between that found for [CuL1]2� (L1 is the anal-ogous macrocycle with ethylenic chains) and that of[CuL3]2� (L3 being the compound with ethylenic chains andN-benzylated pyrazole spacers) (Entry 4 in Table 3). Thesame situation occurs for the dinuclear [Cu2L]4� complexes(Entry 7). The sequence of five-membered chelate rings pre-sent in L1 leads to a stronger coordination to CuII than thesequence of six-membered rings present in L4. The situationfor L3 is different since the substituted spacers cannot bedeprotonated and each one provides only an sp2 nitrogendonor over all the pH ranges studied.

Deprotonation of the pyrazole spacers induced by coor-dination to metal ions is a characteristic feature of thisheterocycle.[8,11,15�17] The structure of L4 enables an analy-sis of the influence of the length of the hydrocarbon chainand thereby of the denticity of the chelate rings on suchdeprotonation processes. The distribution diagrams calcu-lated for an M/L ratio of 2:1 shown in Figure 1 indicatethat while the species [Cu2(H�2L)]2� with both pyrazolemoieties deprotonated is already predominant in solutionat pH � 4.3 for L1, a similar species is produced only at pH


Table 4. Stability constants for the formation of ZnII complexes of the receptors L1, L2, L3 and L4 determined in 0.15 mol·dm�3 NaClat 298.1 K

Entry Reaction[a] L1 L2 L3 L4

1 Zn � L � 3 H � ZnH3L 32.26(7)[b] 32.72(2)2 Zn � L � 2 H � ZnH2L 26.4(2) 27.42(3)3 Zn � L � H � ZnHL 22.26(4) 22.227(9) 16.67(2) 17.54(3)4 Zn � L � ZnL 15.74(4) 14.12(2) 8.53(4) 8.8(1)5 Zn � L � ZnH�1L � H 7.21(4) 3.91(2) �0.43(4) �1.0(1)6 2 Zn � L � H � Zn2HL7 2 Zn � L � Zn2L 20.19(3) 14.40(2) 14.42(3)8 2 Zn � L � Zn2H�1L � H 14.63(5) 14.71(1) 6.04(4) 8.22(1)9 2 Zn � L � Zn2H�2L � 2 H 8.94(4) 4.75(2) �2.71(3) �1.50(3)10 2 Zn � L � Zn2H�3L � 3 H �12.81(3)11 3 Zn � L � Zn3H�3L � 3 H 1.80(4)12 ZnH2L � H � ZnH3L 5.9 5.3013 ZnHL � H � ZnH2L 4.1 5.1914 ZnL � H � ZnHL 6.52 8.11 8.14 8.715 ZnL � ZnH�1L � H �8.53 �10.21 �8.96 �9.816 ZnL�Zn � Zn2L 6.07 5.87 5.6217 Zn2L � H � Zn2HL18 Zn2L � Zn2H�1L � H �5.48 �8.36 �6.219 Zn2H-1L � Zn2H�2L � H �5.69 �9.96 �8.75 �9.7220 Zn2L � Zn2H�2L � 2 H �15.44 �17.11 �15.9221 Zn2H�2L � Zn2H�3L � H

[a] Charges omitted for clarity. [b] Values in parentheses are standard deviations in the last significant figure.

� 8 for L4 (Figure 1, B). The first average deprotonation of[Cu2L4]2� gives pKa1 � 5.1 and the second one pKa2� 8.1.

However, care has to be taken in drawing such a con-clusion since in the red solid red-[Cu2L4](ClO4)4·2H2O{red-[13](ClO4)4·2H2O}, prepared from neutral L4 andCu(ClO4)2·6H2O (vide infra), the two pyrazole units are de-protonated while the central amino groups of both chainsare protonated giving the ligand an overall charge of zero(see 13 in Scheme 3). Therefore, CuII coordination inducesthe ready deprotonation of the pyrazole nitrogen atomswhich, in this complex, occurs at a lower pH than the de-protonation of the central propylenic chain. This is a par-ticular feature of L4 not observed in the related ligand L1

with ethylenic chains. The higher basicity of L4 in its twofirst protonation steps should facilitate this situation. Suchbehaviour contrasts with that of the free ligands, for whichneither deprotonation nor protonation processes occur atthe pyrazole rings throughout the pH range of 2�11.[9,11]

On the other hand, all the dinuclear CuII complexes areEPR-silent.

The stability constants gathered in Table 4 show that, asexpected, the ZnII complexes display remarkably lower sta-bilities than the CuII analogues (see for instance, Entries 4in Table 3 and 4 for the [ML]2� complexes). In all the Zn/L systems, formation of mono- and dinuclear complexescan be observed and for the Zn/L2 system even a trinuclearspecies [Zn3(H�3L2)]4� was detected. The degree of pro-tonation of the mononuclear complexes reaches a value of3 for L1 and L2 {[Zn(H3L)]5� species} and only of 1 forthe systems Zn/L3 and Zn/L4 {i.e. [Zn(HL)]3� species}. Thedegree of protonation of the dinuclear ZnII complexes var-ies between 0 and �3. In the case of the Zn/L1 system,it has not been possible to identify any nondeprotonated


dinuclear species (see distribution diagram in Figure 1). Acomparison with CuII shows that for ZnII, the depro-tonation of pyrazole is shifted approximately two units tohigher pH values as a result of the poorer inductive effectexerted by this metal ion. Taking into account the crystalstructure of [Zn2(H�2L4)](ClO4)2 and the spectroscopicdata (vide infra) it is likely, that in the two species[Zn2(H�1L4)]3� and [Zn2(H�2L4)]2�, both pyrazole groupsare deprotonated. In [Zn2(H�1L4)]3� one of the central ni-trogen atoms of the bridging chain will be protonated. Inthe case of L1 the species [Zn2(H�1L1)]3� cannot be de-tected and the only dinuclear species found in solution is[Zn2(H�2L1)]2� (Figure 1, D and E). For the cryptand L2

the situation is analogous to that of L1 and the species[Zn2(H�1L2)]3� can be postulated as having the two pyra-zole moieties involved in coordination to ZnII being depro-tonated, while the pyrazole in the noncoordinating bridgewill be nondeprotonated and one of the amino groups inthis region will be protonated. Such a situation has alreadybeen observed in the crystal structure of the analogous CuII

complex of L2.[11] Finally, in the ligand with N-benzylatedpyrazole spacers, the main species in solution at neutral pH,is [Zn2L3]4�. Deprotonation of the coordinated water mol-ecules to give hydroxo species requires higher pH values.

Synthesis of Solid ZnII Dinuclear Complexes fromPolyamine Receptors with Ethylenic Chains 1 (L1) to 3 (L3)

Some time ago the in situ deprotonation of ligands 1 (L1)and 2 (L2) upon addition of NaOH, as well as the forma-tion of ZnII pyrazolate complexes was observed using NMRtechniques in [D6]DMSO/D2O mixtures.[18,19] However,


Figure 1. Distribution diagrams for the species existing in equilibrium in the systems: A) CuII/L4 molar ratio 1:1, B) CuII/L4 molar ratio2:1, C) CuII/L1 molar ratio 2:1, D) ZnII/L4 molar ratio 2:1, E) ZnII/L1 molar ratio 2:1; the concentration of ligand is 1 � 10�3 m inall systems

none of the latter mentioned complexes were either isolatedor structurally characterised as solid compounds. However,by treating the neutral ligands L1 to L3 withZn(ClO4)2·6H2O in methanol (molar ratio M/L � 2:1) wehave now isolated three new solid complexes which wererecrystallised from water. The analytical and MS data ofthese compounds are consistent with the formulae[Zn2(H�2L1)](ClO4)2 {[8](ClO4)2}, [Zn2(H�2L2)](ClO4)2·4H2O {[9](ClO4)2·4H2O} and [Zn2L3](ClO4)4·6H2O{[10](ClO4)4·6H2O} (Scheme 3). Electrospray ionisation(ESI-MS) and fast atom bombardment (FAB-MS) massspectrometry have been used for the characterisation of allthe new ZnII complexes.[20] Besides, in the case of[8](ClO4)2, suitable crystals for X-ray analysis were grownfrom the aqueous solution (vide infra).


The above data support the fact that in the case of L1

and L2, ZnII coordination induces the deprotonation of thepyrazole rings to afford the pyrazolate bridging ligand with-out the need to add a base. On the other hand, consideringthat the structure of the starting ligand L3 corresponds toa constitutional isomer in which the 1-benzylpyrazole sub-stituents are in positions opposite each other displaying C2

symmetry[9] and taking into account the crystal structure ofthe analogous CuII complex [7](ClO4)4·4H2O,[11] it isreasonable to assume that in the ZnII complex[10](ClO4)4·6H2O the sp2 nitrogen atoms of the pyrazolerings are also involved in the coordination to the metal ions.

The ESI (positive mode) and FAB mass spectra of com-plex {[8]·(ClO4)2} show common peaks at m/z (%) � 615.1(10) and 615.4 (10), respectively, which confirm the presence


Scheme 3

of the dinuclear [Zn2(H�2L1)(ClO4)]� species, as well as in-tense peaks at m/z (%) � 453.4 (100) and 453.2 (47), respec-tively, corresponding to the species [Zn(H�1L1)]� generatedby loss of Zn(ClO4)2.

Similarly, in both the ESI and FAB mass spectra of[9](ClO4)2·4H2O, the protonated molecular ion[Zn2(H�2L2)(ClO4)2]H� appears at m/z (%) � 893.2 (2) and893.1 (23), respectively, and the ionic species[Zn2(H�2L2)(ClO4)]� is confirmed by the presence of in-tense peaks at m/z (%) � 793.2 (73) and 793.3 (63), respec-tively.

Finally, in the ESI (positive mode) and the FAB massspectra of complex [Zn2L3](ClO4)4·6H2O appear commonpeaks which clearly support the proposed structure. At m/z(%) � 995.1 (5 and 44, respectively) appears a peak corre-sponding to the dinuclear [Zn2L3(ClO4)3]� species formedfrom the molecular ion by loss of perchlorate and at m/z(%) � 733.5 (54) and 733.3 (23), respectively, a peak at-tributable to the ionic species [ZnL3(ClO4)]� formed by suc-cessive loss of Zn(ClO4)2 and ClO4

� can be observed.

Crystal Structure of [Zn2(H�2L1)](ClO4)2 {[8](ClO4)2}

Crystals of [Zn2(H�2L1)](ClO4)2 suitable for X-ray analy-sis were obtained by concentrating aqueous solutions con-taining L1 and Zn(ClO4)2 at pH � 9. The structure consistsof [Zn2(H�2L1)]2� cations and perchlorate anions. EachZnII is coordinated by the three nitrogen atoms from oneof the polyamine bridges and by one sp2 nitrogen atomfrom each one of the pyrazole units which are deprotonated


Figure 2. ORTEP drawing for the cation [Zn2(H�2L1)]2� (8); ther-mal ellipsoids are drawn at the 30% probability level

and behave as bridging η1:η1-bis(monodentate) pyrazolateligands (see Figure 2). The coordination geometry aroundeach one of the metal ions can be defined as midway be-tween a strongly distorted square pyramid and a trigonalbipyramid. The percentage of trigonal-bipyramidal unitscalculated by means of the facial planes[21] is 44.5% for theZn1 site and 50.7% for the Zn2 site.

The shorter distances in the coordination spheres of bothmetal ions are those involving the pyrazolate sp2 nitrogenatoms (see Table 5) and the largest ones are those of theamino groups adjacent to the pyrazole fragments. In order

E. Garcıa-Espana, P. Navarro et al.FULL PAPERto achieve such a coordination geometry the central part ofthe chains are forced to move in the same direction givingsomething of a boat-like shape (Figure 2).

Table 5. Selected bonds lengths [A] and angles [°] of [8](ClO4)2

Parameter Length/angle

Zn1�N1 2.110(5)Zn1�N2 2.212(5)Zn1�N3 2.037(4)Zn1�N9 2.027(4)Zn1�N10 2.267(5)Zn2�N4 2.027(4)Zn2�N5 2.181(5)Zn2�N6 2.109(5)Zn2�N7 2.235(5)Zn2�N8 2.005(4)N1�Zn1�N2 83.3(2)N1�Zn1�N3 110.4(2)N1�Zn1�N9 128.7(2)N1�Zn1�N10 81.1(2)N2�Zn1�N3 78.3(2)N2�Zn1�N10 103.9(2)N3�Zn1�N9 95.4(2)N9�Zn1�N10 76.6(2)N4�Zn2�N5 79.5(2)N4�Zn2�N6 101.6(2)N4�Zn2�N8 94.9(2)N5�Zn2�N6 83.8(2)N5�Zn2�N7 106.8(2)N6�Zn2�N7 82.0(2)N6�Zn2�N8 125.2(2)N7�Zn1�N8 78.3(2)

The Zn�Zn distance (3.99 A) is close to those found forthe analogous CuII complexes of L1 and L2, implying thatthe rigidity of the pyrazole spacer is the main factor con-trolling this parameter. It is interesting to remark that, asalready observed in the systems Cu/L1 and Cu/L2,[11,17] CuII

coordination causes the ready deprotonation of the pyra-zole moieties without requiring any addition of base.

Synthesis of CuII and/or ZnII Dipyrazolate Complexes[Cu2(H�2L4)](ClO4)2 {[11](ClO4)2} and[Zn2(H�2L4)](ClO4)2 {[12](ClO4)2} from 4�

Synthesis and Spectroscopic Characterisation

In order to unequivocally obtain CuII and ZnII dipyrazol-ate complexes we first used the disodium dipyrazolate salt4� (Scheme 2) as a starting material. Treatment of 4� withCu(ClO4)2·6H2O (2 equiv.) in MeOH afforded a blue solidwhich, after being crystallized from a 2:3 water/MeOH mix-ture, gave the CuII dipyrazolate complex [Cu2(H�2L4)]-(ClO4)2 {[11](ClO4)2}.

The ESI (positive mode) mass spectrum of [11](ClO4)2 inMeOH shows two intense peaks at m/z (%) � 671.0 (100)and 571.2 (58) which are assignable to the molecular species[Cu2(H�2L4)(ClO4)]� and [Cu2(H�3L4)]�, respectively.These confirm that the compound is a dinuclear CuII dipyr-


azolate complex. There also appears a fairly weak peak atm/z � 508.1 corresponding to the mononuclear[Cu(H�1L4)]� species which may be formed from the mono-protonated molecular ion [Cu2(H�1L4)(ClO4)2]� by lossof Cu(ClO4)2.

The ZnII dinuclear complex [Zn2(H�2L4)](ClO4)2

{[12](ClO4)2} (Scheme 3) was obtained by treatment of thesodium salt 4� with 2 equiv. of Zn(ClO4)2·6H2O accordingto a similar procedure to that previously mentioned. Aftercrystallisation from a water/methanol mixture, the analyti-cal and spectroscopic data of the white solid obtained wereconsistent with formula [Zn2(H�2L4)](ClO4)2 {[12](ClO4)2}.In its ESI (positive mode) mass spectrum, three main peaksdetected at m/z (%) � 671.2 (32), 609.3 (27) and 509.4 (100)are attributable to the ionic species [Zn2(H�2L4)ClO4]�,[Zn(L4)ClO4]� and [Zn(H�1L4)]�, respectively.

The structure of [Zn2(H�2L4)](ClO4)2 {[12](ClO4)2} insolution was also fully characterised by NMR spectroscopy.In Tables 6 and 7 are gathered the respective 1H and 13CNMR spectroscopic data of this compound recorded in[D6]DMSO together with those of the dipyrazolate salt 4�and those of the free ligand 4. Table 6 verifies that the 1HNMR spectroscopic data of 4 and 4� are similar. Eachshows a single set of signals for all the methylene protons,the chemical shift differences between 4 and 4� being lowerthan 0.2 ppm. However, in the 1H NMR spectrum of[Zn2(H�2L4)]2� (12) (Figure 3, A), the methylene protons6-H2, 8-H2 and 9-H2 which are neighbours of the pyrazole

Table 6. Comparison of the 1H NMR spectra (δ [ppm],[D6]DMSO) of the free ligand 4, its sodium dipyrazolate salt 4�and the ZnII dipyrazolate complex [12](ClO4)2

4 4� [12](ClO4)2

4-H 5.99 (s, 2 H) 5.82 (s, 2 H) 5.77 (s, 2 H)6-HA 3.56 (s, 4 H) 3.56 (s, 4 H) 4.02 (dd, 4 H)[a]

6-HB 3.56 (s, 4 H) 3.56 (s, 4 H) 3.50 (dd, 4 H)[b]

8-HA 2.48 (m, 4 H) 2.53 (m, 4 H) 3.09 (m, 4 H)8-HB 2.48 (m, 4 H) 2.53 (m, 4 H) 2.71 (m, 4 H)9-HA 1.49 (m, 4 H) 1.49 (m, 4 H) 1.93 (m, 4 H)9-HB 1.49 (m, 4 H) 1.49 (m, 4 H) 1.69 (m, 4 H)10-H 2.48 (m, 8 H) 2.47 (m, 8 H) 2.96 (m, 8 H)7-H � � 4.77 [m (t), 4H]11-H � � 5.03 [m (t), 2H]

[a] 1J � 16 Hz; 2J � 7 Hz. [b] 1J � 16 Hz; 2J � 3 Hz.

Table 7. Comparison of the 13C NMR spectra (δ [ppm],[D6]DMSO) of the free ligand 4, its sodium dipyrazolate salt 4�and the ZnII dipyrazolate complex [12](ClO4)2

4 (1) 4� (2) ∆δ (2�1) {[12](ClO4)2} (3) ∆δ (3�2)

C-3,5 146.97[a] 147.31[b] 0.34 151.12[b] 3.81C-4 101.71 100.58 �1.13 96.88 �3.70C-6 45.27 46.26 0.99 45.90 �0.36C-8 47.02 47.14 0.12 50.67 3.56C-9 29.22 29.26 0.04 23.72 �5.94C-10 47.67 47.78 0.11 49.96 2.18

[a] Broad signal. [b] Sharp singlet.


Figure 3. 1H (A) and 13C (B) NMR spectra of the ZnII complex{[12](ClO4)2} in [D6]DMSO solution

moieties, appear in the spectrum as an AB spin system (Fig-ure 3) suggesting that ZnII coordination confers a rigidstructure on the complex. Except for the 4-H pyrazole pro-ton, all the other protons belonging to the macrocyclic cav-ity of 12 exhibit a generalised downfield chemical shift oftheir signals. It is interesting to note that in the 1H NMRspectrum appear two signals at δ � 4.77 and δ � 5.03 ppm,which are attributable to the protons of the amino groups7-H and 11-H, respectively. The triplet-like shape of thesesignals can be ascribed to their coupling with the neighbourmethylene groups.

Table 7 shows the 13C NMR signals of 4, 4� and[12](ClO4)2 as well as the variations in the chemical shiftsdue to deprotonation of the pyrazole moieties which leadsto 4� [∆δ (2�1)] and the formation of the dinuclear ZnII

complex 12 [∆δ (3�2)].In [D6]DMSO, the C-3,5 quaternary carbon atoms of the

free ligand 4 appear in the spectrum as a broad signal, whilein the spectrum of 4� this signal changes to a sharp singletindicating that the slow prototropic equilibrium of the pyra-zole rings has disappeared. Furthermore, as can be ex-pected, upfield (δ �1.13 ppm for C-4) and downfield (δ �0.34 ppm for C-3,5) variations of the pyrazole carbon sig-nals confirm the formation of pyrazolate anions (Table 7).


Taking the 13C NMR signals of 4� as a reference, it canbe observed that formation of the ZnII dipyrazolate com-plex 12 induces strong downfield shifts in the signals of allthe carbon atoms located in the α- positions to the nitrogenatoms involved in the complexation of the ZnII ions. It alsoexerts strong upfield shifts on the signals of those locatedin the β- positions (Table 7).

Crystal Structure of [Cu2(H�2L4)](ClO4)2 {[11](ClO4)2}

The crystallographic data were recorded at 150 K be-cause some disorder found within the propylenic chains ofthe bridges and in the perchlorate counter anions preventeda satisfactory solution of the crystal structure from datacollected at room temperature. Although even at low tem-perature some disorder persists, the solution is good enoughfor unequivocally identifying the structural characteristicsof the coordination sites.

Crystals of [11](ClO4)2 consist of [Cu2(H�2L4)]2� cationsand perchlorate anions. The coordination geometry aroundeach copper atom is an irregular, axially elongated squarepyramid. The base of the pyramid is formed by the twosecondary nitrogen atoms of the polyamine bridge closer tothe heterocycle and by two nitrogen atoms of two differentpyrazolate units which act as η1:η1 bis(monodentate)anionic bridging ligands (Figure 4). The displacements ofthe metal ions from the plane of the nitrogen donors are0.322(2) A for Cu1 and 0.303(2) A for Cu2. The centralnitrogen atoms of the bridges occupy the axial positions. Inboth coordination sites, the Cu�N distances involving thesp2 pyrazolate nitrogen atoms [Cu(1)�N(3) � 1.963(9) A,Cu(2)�N(4) � 1.94(1) A; Table 8] are much shorter thanthose of the secondary nitrogen atoms [Cu(1)�N(2) �2.10(1) A, Cu(2)�N(5) � 2.11(1) A]. The Cu(1)�Cu(2)distance is 3.96 A which is almost identical to thosefound in the crystal structures of [Cu2(H�2L1)](ClO4)2

{[5](ClO4)2}[17,11] and [Cu2(H�1L2)](ClO4)3·2H2O{[6](ClO4)2(HClO4)·2H2O}[11] as well as in several crystalstructures of other pyrazole-containing complexes in whichthe pyrazolate anions exhibit the same coordinationmode.[8,15,16] However, such a distance is not so fixed insome complexes of podand ligands containing pyrazolebridges.[22] It is also interesting to point out that the axial

Figure 4. ORTEP drawing for the cation [Cu2(H�2L4)]2� (11); ther-mal ellipsoids are drawn at the 30% probability level



Length/angleParameter

Cu1�N1 2.20(2)Cu1�N2 2.1(1)Cu1�N3 1.963(9)Cu2�N4 1.94(1)Cu2�N5 2.11(1)Cu2�N6 2.24(2)N1�Cu1�N2 92.6(5)N1�Cu1�N3 106.8(5)N2�Cu1�N3 80.5(4)N4�Cu2�N5 81.1(4)N4�Cu2�N6 106.4(4)N5�Cu2�N6 91.7(5)

elongation observed in this complex is much smaller thanthose previously reported for [Cu2(H�2L1)](ClO4)2 and[Cu2(H�1L2)](ClO4)3·2H2O.[11] The complex again adopts aboat-shaped conformation with the side-chains foldedtowards the same side of the macrocyclic cavity.

Crystal Structure of [Zn2(H�2L4)](ClO4)2 {[12](ClO4)2}

The crystallographic data were also recorded at 150 Kbecause similar disorder to that found in the previous casewas observed and this prevented a satisfactory solutionfrom data recorded at room temperature. Crystals of[12](ClO4)2 consist of [Zn2(H�2L4)]2� cations and perchlor-ate anions. The coordination geometry around each ZnII

atom is very irregular and can be defined as a distortedsquare pyramid. Analogous to the previous structure, thebase of the pyramid is formed by the two secondary nitro-gen atoms of the polyamine bridge closer to the heterocycleand by two nitrogen atoms of two different pyrazolate unitswhich act as η1:η1 bis(monodentate) anionic bridging li-gands (Figure 5). The displacements of the metal atomsabove the plane defined by the nitrogen donors are 0.595(2)A for Zn1 and 0.554(2) A for Zn2. The central nitrogenatoms of the polyamine bridges occupy the axial positionswhich in this case do not show any distortion. In contrastto the analogous CuII complex, the Zn�N distances of thesp2 pyrazole nitrogen atoms in this case [Zn(1)�N(3) �


Length/angleParameter

Zn1�N1 2.08(1)Zn1�N2 2.206(8)Zn1�N3 2.037(9)Zn2�N4 2.040(8)Zn2�N5 2.196(8)Zn2�N6 2.09(1)N1�Zn1�N2 99.5(4)N1�Zn1�N3 114.3(4)N2�Zn1�N3 77.4(3)N4�Zn2�N5 77.38(3)N4�Zn2�N6 112.7(3)N5�Zn2�N6 98.1(3)


2.037(9) A, Zn(2)�N(4) � 2.040(8) A; Table 9] are close tothose of their secondary nitrogen atoms in the centre of thebridge [Zn(1)�N(1) � 2.08(1) A, Zn(1)�N(2) � 2.09(1) A].The largest M�N distances are found for the secondarynitrogen atom adjacent to the heterocycle. TheZn(1)�Zn(2) distance is again 3.96 A confirming that it isfixed by the pyrazolate bridge. The conformation of thewhole molecule is again boat-shaped with the central partof both chains folded towards the same side of the macro-cyclic plane (Figure 5).

Figure 5. ORTEP drawing for the cation [Zn2(H�2L4)]2� (12); ther-mal ellipsoids are drawn at the 30% probability level

Synthesis of CuII Dipyrazolate Complexes[Cu2(L4)](ClO4)4·2H2O {[13](ClO4)4·2H2O} and [Cu3(H�2L4)](ClO4)4·2MeOH {[14](ClO4)4·2MeOH} from L4

Synthesis and Spectroscopic Characterisation

It is interesting to point out that when we tried toobtain the dipyrazolate complexes [Cu2(H�2L4)](ClO4)2

{[11](ClO4)2} by direct treatment of the neutral ligand 4with Cu(ClO4)2·6H2O in methanol, we observed very differ-ent behaviour to that previously described using the disod-ium dipyrazolate salt 4� as a starting material. Thus, treat-ment of a methanolic solution of 4 (2.5 � 10�2 m) withCu(ClO4)2·6H2O (2 equiv.) at room temperature gave a blueprecipitate which, after being vacuum-dried at 100 °C, af-forded a blue solid of formula [Cu2L4](ClO4)4·2H2O in 71%yield. After redissolving this blue solid in boiling water for5 min, a red solid slowly and unexpectedly crystallised.After being vacuum-dried at 100 °C, this resulted in a red-dish-brown complex of formula [Cu2L4](ClO4)4·2H2O{[13](ClO4)4·2H2O} (Scheme 3) in 61% yield. Surprisingly,this red complex had identical analytical data to the blueprecursor mentioned above. Microanalysis performed by X-ray energy dispersive spectrometry confirmed the proposedCu/Cl ratio. This can be explained if both the blue and thered forms of this complex correspond to dinuclear CuII di-pyrazolate species which share the characteristic feature ofhaving the two sp3 nitrogen atoms located at the centre ofthe side chains protonated and, in contrast to[Cu2(H�2L4)](ClO4)2 {[11](ClO4)2}, thereby not directly co-ordinated to the metal atoms (Scheme 3). The coordination

CuII and ZnII Coordination Chemistry of Pyrazole-Containing Polyamine Receptors FULL PAPERof the metal ion causes the release of the protons from thepyrazole fragments which are captured by the basic centralnitrogen atoms of the bridges.

Furthermore, the ESI-MS data of either the precursorblue form or the red final form show identical peaks whichappear with different relative abundances as shown inTable 10. In the negative mode, the molecular ion [M �H]�, corresponding to [Cu2(H�1L4)(ClO4)4]� appears atm/z � 969.4 with low relative abundance (see Table 10), to-gether with a more intense peak at m/z � 1071.4 and thebase peak at m/z � 869.4 assignable to the species[M � ClO4]� and [(M � ClO4 � 2 H]�, respectively,which agree with the proposed structure. Moreover, in thepositive mode, two characteristic peaks appear at m/z � 871and 771, assignable to the [Cu2(L4)(ClO4)3]� and[Cu2(H�1L4)(ClO4)2]� species which correspond to di-nuclear CuII dipyrazolate fragments containing partly pro-tonated polyamine chains. On the other hand, in additionto the expected peak at m/z � 671 assignable to the mostabundant fragment [(H�2L4)Cu2(ClO4)]�, both the precur-sor blue and the final red forms of [13](ClO4)2(H2O)2 showa characteristic peak at m/z � 547 attributable to[(H2L4)(ClO4)]� which demonstrates the presence of pro-tonated nitrogen atoms in the ligand structure.

Table 10. Comparison of the most significant peaks recorded in theESI-MS data of the dinuclear CuII complex {[13](ClO4)4·2(H2O)}(1007.6) in both its blue and red forms

m/z Composition % blue % red

Negative mode1071 {[Cu2L4](ClO4)5}� 35 14969 {[Cu2(H�1L4)](ClO4)4}� 6 8869 {[Cu2(H�2L4)](ClO4)3}� 100 100Positive mode871 {[Cu2L4](ClO4)3}� 10 3771 {[Cu2(H�1L4)](ClO4)2}� 10 4671 {[Cu2(H�2L4)](ClO4)}� 100 59569 [Cu2(H�3L4)]� 3 4547 {[H2L4](ClO4)}� 51 33447 [HL4]� � 41224 [(L4/2)�H]� � 100

Unfortunately, up to now, it has not been possible to ob-tain adequate crystals of either the precursor blue form orthe red form suitable for X-ray analysis.

However, on the basis of the above observations, the col-our changes observed can be tentatively attributed to thepresence of the two water molecules which are bound in adifferent manner in both complexes. In the blue form theywould be coordinated to the metal ion occupying the axialpositions of the pentacoordinate environment (seeScheme 4, blue-[13]). In the red form the coordination en-vironment around each CuII will be square-planar and thewater molecules would not occupy the first coordinationsphere of the metal ion (Scheme 4, red-[13]). In both casesthe water molecules might be stabilised through�NH···OH2 hydrogen bonds with the two protonated sp3


nitrogen atoms located in the centre of the side chains. Wepreviously reported these types of monohydrated am-monium groups in the crystalline structure of the dinuclearCuII complex [6](HClO4)(ClO4)2·2H2O[11] where one of thealiphatic nitrogen atoms of the noncoordinated bridge isprotonated and hydrogen-bonded to a water molecule.

Scheme 4

In the complex {red-[13](ClO4)2·2H2O}, the confor-mational change produced would be induced by the treat-ment of the starting precursor blue form with boiling water.

Finally, when we tried to obtain similar dinuclear com-plexes starting from a more concentrated solution of ligandL4 (4) (6.0 �10�2 m) under similar reaction conditions [2equiv. of Cu(ClO4)2·6H2O at room temp.], we obtained an-other unexpected result. A bright green solid precipitatedfrom the methanol solution which, after being vacuum-dried at 100 °C, gave a new trinuclear copper complex inlow yield (28%). The analytical and MS data of this newcomplex are consistent with the formula[Cu3(H�2L4)](ClO4)4·2MeOH {[14](ClO4)4·2MeOH}

E. Garcıa-Espana, P. Navarro et al.FULL PAPER(Scheme 3). The Cu/Cl (3:4) molar ratio was confirmed byX-ray energy dispersive spectrometry.

Since matrix-assisted laser desorption/ionisation(MALDI-TOF) mass spectrometry is actually among themost important ionisation methods for strong metal-ioncomplexes,[23] we used this technique in order to confirmthe structure of the trinuclear complex. Table 11 shows themost significant trinuclear copper ions detected in theMALDI-TOF mass spectrum of {[14](ClO4)4·2MeOH}using 2,5-dihydroxybenzoic acid (DHB) as a matrix. In asimilar way as has already been observed in other MALDI-TOF mass spectra of copper complexes in which CuII ionsare apparently simply reduced,[24,25] the composition ofmost of the ionic fragments gathered in Table 11 agree withdifferent copper oxidation states due to partial or total cop-per reduction processes for which the source of the electronsmay be the photoionised DHB matrix. Thus, at m/z � 635appears the base peak corresponding to the skeletal trinu-clear copper fragment [Cu3(H�2L4)]� in which the threeCuII atoms are reduced to CuI. Other characteristic ionswhich correspond to the proposed structure appear at m/z(%) � 734.3 (78) and 797.2 (27). Both peaks are attributableto the ionic species [Cu3(H�2L4)(ClO4)]� and[Cu3(H�2L4)(MeO)(MeOH)]�, respectively, in which two ofthe three CuII atoms have been reduced to CuI. The lastmentioned species (m/z � 797.2) confirms the presence oftwo MeOH molecules in the complex and simultaneouslysuggests that one of these molecules may be acting as abridge between two copper ions.

Table 11. Most significant peaks recorded in the MALDI-TOF-MS of [14](ClO4)4·2MeOH corresponding to trinuclear copperfragments

m/z (%) Composition[a] Oxidation state

797.2 (27) {[Cu3(H�2L)](MeO)(MeOH)}� 2 Cu� � Cu2�

750.3 (7) {[Cu3(H�2L)](ClO4)�(O)}� 3 Cu2�

734.3 (78) {[Cu3(H�2L)](ClO4)}� 2 Cu� � Cu2�

649.4 (26) {[Cu3(H�4L) �(O)]}·� 3 Cu2�

635.4 (100) [Cu3(H�2L)]� 3 Cu�

[a] [(H�2L) � (C22H40N10) and (H�4L)�(C22H38N10)].

Finally, at m/z � 750.3 and 649.4 are two peaks assign-able to ionic fragments containing oxygen in which the CuII

initial oxidation state has not changed. However, it is diffi-cult to be certain if these ions are truly formed or are theresult of secondary oxidation reactions in the matrix as waspreviously observed in other MALDI-TOF mass spectra ofmetallic complexes.[25]

Molecular Modelling Studies

In order to gain further insight into the molecular struc-tures of both the dinuclear CuII [13](ClO4)4·2H2O and thetrinuclear CuII complex {[14](ClO4)4·2MeOH}, we per-formed some modelling studies. Molecular mechanicsmodels of complexes of a variety of metal ions have been


recently developed[26] and have proved useful for ligand de-sign. Use of a well-established biochemically oriented forcefield with specific metal parameters would allow the inter-actions of metal complexes and biological molecules to bestudied and modelled. Among the well-established force-fi-elds, AMBER[27] has found widespread use in the modellingof biological molecules and the development of suitableparameter sets which allows the modelling of metal com-plexes. This has produced results in good agreement withexperimental observations without requiring additionalchanges.[28] For those reasons, in this work, molecular mod-elling studies were carried out using the AMBER methodmodified by the inclusion of appropriate parameters.

To check the consistency of the calculations we modelledthe complex [11](ClO4)2 (which has a known X-ray struc-ture) according to the same procedure detailed in the Exp.Sect. with no restraints. Data obtained in this modellingstudy agree quite well with the crystallographic results, al-though the calculated Cu�Cu distances are shorter thanthe experimentally measured ones (0.12 A, similar to maxi-mum deviation obtained by Comba et al.).[29]

To study the copper complexes [13](ClO4)4·2H2O and[14](ClO4)4·MeOH, coordination numbers and geometrieswere selected according to the preferences of copper(ii) and,taking into account all the experimental data for the syn-thesised complexes, red-[13] was assumed to be square-planar while blue-[13] and green [14] were assumed to bepyramidal. Thus, we modelled the blue and red forms ofthe dihydrated complex [13](ClO4)4·2H2O with the resultsgiven in Figure 6 (A).

Figure 6. A) Molecular models of the blue and red forms of thecomplex [13](ClO4)4(H2O)2; B) molecular model for the trinuclearcomplex [14](ClO4)4(MeOH)2

In the above complexes, the calculated Cu�Cu distancebetween the two metal ions coordinated with pyrazole ni-

CuII and ZnII Coordination Chemistry of Pyrazole-Containing Polyamine Receptors FULL PAPERtrogen atoms is 3.52 A for the red form of [13], denotingthe rigidity of the bridge, and 4.03A for its blue form whichis more similar to the values found for [11] (calculated 3.84A, X-ray data 3.96 A).

Since they have different connectivities, no attempt wasmade to compare the calculated molecular mechanics strainenergies of the different coordination complexes since eachone needs a different parameter set. PM3 calculations ofthe corresponding heats of formation, however, suggeststhat the red form of {[13](ClO4)4·2H2O} is more stable thanthe blue one.

Finally, a model for the green trinuclear CuII complex[14](ClO4)4·2MeOH is given in Figure 6 (B). Coordinationof the metal centres to the pyrazole sp2 nitrogen atoms andthe adjacent nitrogen atoms forces the central parts of bothbridges to move upwards in the same direction thereby en-abling the central nitrogen atoms to coordinate to a furthermetal ion provided there are other ancillary ligands that canassist the processes. In this case the methanol moleculescould complete the coordination spheres. In [14], the calcu-lated Cu�Cu distances between the third metal ion and thepyrazole-coordinated metal ion are 5.89 A and 6.41 A,respectively, while the distance between these two remainsquite similar to that of 3.83 A in the blue form of [13].

1H NMR Spectra of {red-[13](ClO4)4·2H2O}

In order to obtain further information about the natureof the red complex [Cu2L4](ClO4)4·2H2O {red-[13](ClO4)4·2H2O} (Scheme 3), we performed paramagnetic1H NMR spectroscopy. This technique is very useful forthe study of magnetically coupled systems such as dinuclearCuII�CuII complexes.[30,31] However, in the case of mono-nuclear CuII complexes its usefulness is rather limited dueto the large electronic relaxation time of CuII which conse-quently produces a large broadening of the signals. The in-formation that this technique provides refers to the elec-tronic properties, coordination geometry and confor-mational changes in solution.

The 1H NMR spectrum of the red [Cu2L4](ClO4)4·2H2Ocomplex in (CD3)2SO is shown in Figure 7 (A). It displays,

Table 12. 1H NMR hyperfine-shifted resonances of red-[13](ClO4)4·2H2O in (CD3)2SO at 298 K

δ [ppm] Assignments Number of protons T1 [ms] ∆ν1/2 [Hz] T2 [ms][a]Signal

a 42.3 α-CH2 � 1 1740 0.18b 15.8 α-CH2 16 � 1 2128 0.15c 14.2 α-CH2 � 1 [b] [b]

d 8.2 Hpyrazole[c] 1 138 52[d] 6.1[d]

e 7.8 γ-CH2 8 108 [b] [b]

f 7.1 Hpyrazole[c] 1 146 [b] [b]

g 6.9 β-CH2 146 [b] [b]

h 6.4 β-CH2 150 [b] [b]

i 6.2 β-CH2 152 [b] [b]

j 0.27 β-CH2 8 130 [b] [b]

k 0.05 β-CH2 158 [b] [b]

l �0.32 β-CH2 158 65[d] 4.9[d]

m �1.4 β-CH2 114 322[d] 1.0[d]

[a] Measured from the line width at half-height. [b] Overlap prevents measurement of this value. [c] Tentative assignments. [d] Measured at323 K (blue complex).


in the downfield region, nine hyperfine-shifted signals (a�i)and four other signals (j�m) which are shifted upfield.When the temperature is raised up to 45 °C, the broad sig-nals a�c and e move upfield, Figure 7 (B, C). This changeis irreversible and at the same time the colour of solutionchanges from red-violet to blue, indicating a confor-mational change. The isotropically shifted resonances,longitudinal relaxation times (T1) and line widths at half-height are reported in Table 12 for the red complex[Cu2L4](ClO4)4·2H2O {red-[13](ClO4)4·2H2O}. The above-mentioned signals (a�c) display short T1 values (� 1 ms)and broad line widths at half-height of around 2000 Hz.This, together with the proton integrations, permits the as-signment of these signals to the protons of the CH2 groupslocated α to the coordinated nitrogen atoms nearest to theCuII ion (see Scheme 5). When the temperature is raised upto 45 °C (see Figure 7, C), the broad signals which moveupfield integrate for eight more protons, indicating an in-crease in the coordination number of the copper ion. In thissense, signal (e) which integrates for eight protons and can

Figure 7. 400 MHz 1H NMR spectra of [Cu2L4)](ClO4)4(H2O)2 in(CD3)2SO at different temperatures: 298 K (A); 313 K (B); 323 K(C); the asterisks mark the residual solvent and impurity signals [*(CH3)2SO; ** HOD]

E. Garcıa-Espana, P. Navarro et al.FULL PAPERbe assigned to CH2 groups located γ to the coordinatednitrogen atoms in the red-violet [Cu2L4](ClO4)4·2H2O {red-[13](ClO4)4·2H2O} complex, disappears completely in the1H NMR spectrum of the blue complex 13. These resultsclearly illustrate a change in the coordination geometryfrom square-planar in the red form to pyramidal or five-coordinate in the blue form. The data also suggest that inthe blue form, the nitrogen atoms in the centre of thebridges are coordinated to the metal ion. Therefore, in theweakly acidic solvent (CD3)2SO, the centre nitrogen atomsare deprotonated and the red complex gradually transformsto [Cu2(H�2L4)]2� (11) (see Scheme 5). Indeed, this spec-trum is identical to that obtained of a solution formed bydirectly dissolving [Cu2(H�2L4)](ClO4)2 ([11](ClO4)2] in thesame solvent. The other signals (d and f�m), which corre-spond to the CH2 groups β to the coordinated nitrogenatoms and to the pyrazole protons, display large T1 values(100�150 ms). As described previously,[31] dinuclear CuII

complexes have similar T1 values and broad line widths.

Scheme 5

In addition, variable-temperature 1H NMR spectra ofred-[Cu2L4](ClO4)4·2H2O were recorded from 293 to 323 K.The representation of the temperature dependence of theparamagnetic hyperfine-shifted resonances for some pro-tons is shown in Figure 8. All signals are temperature-inde-pendent except for signals (a) and (b). Only signal (a) exhib-its anti-Curie behaviour and furthermore signal (b) showsCurie behaviour. Extrapolated values at infinite tempera-tures for signal (a) indicate the existence of magnetic coup-ling.[31]


Figure 8. Temperature dependence of the isotropically shifted res-onances of red-[13](ClO4)4·2H2O; the insert shows the plot with the1/T coordinate expanded

Magnetic Measurements of {red-[13] (ClO4)4·2H2O} and{[11] (ClO4)2}

The magnetic properties of [Cu2(H�2L4)](ClO4)2

{[11](ClO4)2} and [Cu2(L4)](ClO4)4·2H2O {red-[13](ClO4)4·2H2O} are shown in Figure 9 in the form of χMTversus T, χM being the magnetic susceptibility per twocopper(ii) ions. The value of χMT at room temperature is0.46 cm3·mol�1 K for [Cu2L4](ClO4)4·2H2O and[Cu2(H�2L4)](ClO4)2, a value which is significantly lowerthan that expected for two uncoupled CuII ions. Upon cool-

Figure 9. Plot of χMT vs. T for the complexes red-[13](ClO4)4·2H2Oand [11](ClO4)2

CuII and ZnII Coordination Chemistry of Pyrazole-Containing Polyamine Receptors FULL PAPERing χMT continuously decreases, vanishing at around 70 K.Moreover, the corresponding χM curves show maxima at270 K (for [Cu2L4](ClO4)4·2H2O) and 255 K (for[Cu2(H�2L4)](ClO4)2] (see Supporting Information, FigureS2). This magnetic behaviour is characteristic of an import-ant antiferromagnetic interaction between the CuII ionswith singlet spin ground states.

We have interpreted the magnetic data according to theactual dinuclear nature of the complexes with the spinHamiltonian H � �JS1S2. The fit of the magnetic data tothe appropriate susceptibility expression derived from thevan Vleck formula and assuming that the g factors for bothCuII ions are identical, leads to J � �299(7) cm�1 andg � 2.09(1) for red-[Cu2L4](ClO4)4·2H2O {red-[13](ClO4)2·2H2O}, and J � �286(6) cm�1 and g � 2.07(1)for [Cu2(H�2L4)](ClO4)2 {[11](ClO4)2}.

The observed J values lie in the upper range of the mag-netic interactions found for related bis(µ-pyrazolato)-bridged dinuclear complexes. In fact, to the best of ourknowledge, the highest J value observed for this family isJ � �412 cm�1 for the compound [Cu2L2](BF4)2 whereL � 3,5-bis(iPrSCH2)pyz.[32] The existence of σ-donoratoms as peripheral ligands may be responsible for thesestrong interactions.[33] However, the number of structurallycharacterised examples reported in the literature is still lim-ited.[34]

The pathway for magnetic exchange is propagatedthrough the bridging pyrazolate ligands. From the molecu-lar structure of [Cu2(H�2L4)](ClO4)2 {[11](ClO4)2} onecan conclude that the unpaired electron in each metalcentre is clearly described by a dx

2�y

2 magnetic orbitalwhich is coplanar with the pyrazolate skeleton. The signifi-cant overlap between these magnetic orbitals accounts forthe strong antiferromagnetic coupling observed. Althoughthe crystal structure of [Cu2L4](ClO4)4·2H2O is notavailable, the very close magnetic behaviour of the com-plexes [Cu2L4](ClO4)4·2H2O {red-[13](ClO4)4·2H2O} and[Cu2(H�2L4)](ClO4)2 allows us to predict a similar exchangepathway. This again supports the suggestion that depro-tonation of the pyrazole rings has occurred in both struc-tures.

Conclusions

The synthesis of a 2�2 polyamine macrocycle containingpyrazole spacers connected through methylene groups to di-propylenetriamine bridges (L4) has been described here forthe first time. The acid-base behaviour of L4 in water re-veals that, like the analogous macrocycle containing dieth-ylenetriamine bridges (L1), the pyrazole spacers do not par-ticipate in any net protonation step in the pH range of2�11. The number of protonation steps observed is equalto the number of amino groups in the bridges. 1H and 13CNMR spectroscopy support this conclusion. The overallbasicity of L4 is, on the other hand, five orders of magni-tude greater than that of L1. The deprotonation of the pyra-zole moieties can, however, be readily achieved in the pres-


ence of coordinated metal ions. CuII induces the depro-tonation of the pyrazole rings at lower pH values than ZnII

due to its much stronger interaction with the ligand. On theother hand, pyrazole deprotonation is more readilyachieved (at more acidic pH values) with the ligand L1 bear-ing ethylenic chains in the bridges with five-membered che-lates than in L4 containing six-membered chelates and ahigher overall basicity.

The higher basicity of L4 is also reflected in the differentsyntheses of the dinuclear solid complexes which we havecarried out. In the case of L1, treatment of Cu(ClO4)2·6H2Owith either neutral L1 or with its disodium salt Na2[H�2L1]in ethanol followed by crystallisation from water leads tothe formation of the same dinuclear complex[Cu2(H�2L1)](ClO4)2 in which the pyrazole groups are de-protonated thereby behaving as η1:η1 monodentate bridg-ing ligands and the polyamine bridges are completely de-protonated. In the case of L4 the situation is different.Treatment of neutral L4 with Cu(ClO4)2·6H2O in MeOHgives a blue solid complex which by heating in boiling water(for 5 min) followed by slow crystallisation leads to a redcompound. The elemental analyses of both the blue andthe red forms of this complex are consistent with the sameempirical formula, i.e. [Cu2L4](ClO4)4·2(H2O). The ESI-MSand 1H NMR paramagnetic data of the red complex clearlyindicates that it is a dinuclear square-planar CuII complexin which the central nitrogen atoms of the bridging chainsare not involved in the coordination to the metal ions andare therefore protonated. On the basis of both the elementalanalysis of {red-[13](ClO4)4·2H2O} and crystallographicdata of polyamine receptors previously reported, it seemsreasonable to suppose that the above-mentioned protonatednitrogen atoms may be hydrogen-bonded to water mol-ecules [Scheme 4, 13 (red form)]. The blue form shows thesame feature but in this case the coordination geometryshould be pyramidal with the water molecules occupyingthe axial positions. The exchange between the two formsdepends on both the temperature and the solvent. There-fore, the marked basicity of the propylamine chains facili-tates the capture of the protons released from the pyrazolerings.

However, if Cu(ClO4)2·6H2O is treated with the sodiumsalt Na2[H�2L4] in methanol followed by crystallisationfrom water, the complex obtained is [Cu2(H�2L4)](ClO4)2

{[11](ClO4)2} which is analogous to that obtained with L1

(Scheme 3). The crystal structure of [Cu2(H�2L4)](ClO4)2

shows a very irregular square-pyramidal coordination ge-ometry in which the shorter distances are those involvingthe pyrazole nitrogen atoms and the axial ones are involvedin both coordination sites by the central nitrogen atoms ofthe propylenic chains. Analogously, in the case of ZnII, thecrystal structures of [Zn2(H�2L1)(ClO4)2 {[8](ClO4)2} and[Zn2(H�2L4)](ClO4)2 {[12](ClO4)2} also reveal similar coor-dination arrangements. In the three structures, the overallshape of the complex cations is boat-like with the centralparts of the polyamine bridges displaced towards the sameside of the macrocyclic ring.

E. Garcıa-Espana, P. Navarro et al.FULL PAPERFinally, treatment of concentrated solutions of L4 with

Cu(ClO4)2·6H2O gives a trinuclear complex of stoichi-ometry [Cu3(H�2L4)](ClO4)4·2MeOH in low yield whichhas been characterised by MALDI-TOF mass spec-trometry. The boat-like arrangement of the macrocycle inthe dinuclear complex seems to facilitate the process.

Experimental Section

General: All starting materials were purchased from commercialsources (Aldrich) and used without further purification. Solventswere dried using standard techniques.[35] The organic reactionswere monitored by thin layer chromatography (TLC) on precoatedaluminium sheets of silica gel. Compounds were detected with iod-ine and/or with phosphomolybdic acid. Flash column chromatog-raphy was performed in the indicated solvent system on silica gel(particle size 0.040�0.063 mm). Melting points were determinedwith a Reichert-Jung hot-stage microscope. 1H and 13C NMR spec-tra were recorded with Varian Unity Inova-300, Varian Unity In-ova-400 or Varian Unity 500 spectrometers at room temperature,employing D2O or [D6]DMSO as solvents. The pH was calculatedfrom the measured pD values using the correlation pH � pD �

0.4.[36] Chemical shifts are reported in parts per million (ppm) fromtetramethylsilane but were measured against the solvent signals. Di-oxane (δ � 67.4 ppm) was used as reference for the 13C NMR spec-tra in D2O. All assignments were performed on the basis of 1H-13Cheteronuclear multiple quantum coherence experiments (gHSQCand gHMBC). IR spectra were recorded with a Perkin�Elmer 681spectrometer. EI mass spectra were recorded with a Hewlett Pack-ard 5973 spectrometer with a direct insertion probe (70 eV). Elec-trospray mass spectra were obtained with a Hewlett Packard 1100MSD and FAB mass spectra with a VG AutoSpec spectrometerusing an m-nitrobenzyl alcohol (NBA) matrix. Elemental analyseswere provided by the Departamento de Analisis, Centro de Quım-ica Organica ‘‘Manuel Lora Tamayo’’, CSIC, Madrid, Spain.Cu�Cl ratios were provided by the SCSIE (University of Valencia)using X-ray energy dispersive spectroscopy with an EDAX DX4detector and a Philips XL30 ESEM electronic microscope.

Synthesis of Ligands: Ligands L1 (1), L2 (2) and L3 (3) were pre-pared by treatment of the corresponding 3,5-pyrazoledicarbal-dehydes with diethylenetriamine [L1 (1) and L3 (3)] or tris(2-amino-ethyl)amine L2 (2) as reported previously in refs.[10,11], respectively.

3,7,11,14,15,18,22,26,29,30-Decaazatricyclo[26.2.1.113,16]dotri-aconta-1(31),13,16(32),28-tetraene [L4 (4)]: 3,5-Pyrazoledicarbal-dehyde (496 mg, 4.0 mmol) was dissolved in hot dry methanol(200 mL). The solution was then cooled to room temperature andadded dropwise under argon to a stirred solution of N-(3-amino-propyl)-1,3-propanediamine (525 mg, 4 mmol) in dry methanol(120 mL) over 2.5 h. The reaction was monitored by TLC (CHCl3/MeOH, 10:1) and when it was complete (ca. 12 h), NaBH4 (912 mg,24.0 mmol) was added portionwise over 15 min. The mixture wasstirred for 2 h and the solvent removed in vacuo. The residue waspurified by flash column chromatography (MeOH/30% aqueousNH4OH, 49:1 to 45:5 mixtures). The fractions containing the prod-uct of Rf � 0.65 (TLC, MeOH/30% aqueous NH4OH, 1:1) wereconcentrated to dryness. The residue was then extracted with boil-ing toluene and on concentration and cooling gave L4 (4) as an oilwhich was crystallised from toluene to give a white solid (443 mg,49%) m.p. 154�156 °C. IR (KBr): ν � 3500�3300 cm�1 (NH).MS (FAB�, 3-nitrobenzyl alcohol matrix): m/z (%) � 447 (100)


[MH�]. C22H42N10 (446.6): calcd. C 59.16, H 9.48, N 31.36; foundC 58.92, H 9.62, N 31.11.

Preparation of L4·6HCl (4·6HCl): A mixture of L4 (4) (223 mg,0.5 mmol) and 1 m aqueous hydrochloric acid (4 mL) was stirredfor 12 h. After addition of EtOH (50 mL), a white precipitateformed which was filtered off and dried under reduced pressure for48 h to give a white solid (319 mg, 91%); m.p. 265�268 °C. MS(ESI positive mode, H2O): m/z (%) � 447 (100) [(M � 6 HCl)H�].C22H42N10·6HCl·2H2O (701.4): calcd. C 37.67, H 7.47, N 19.97;found C 38.05, H 7.70, N 19.78.

Preparation of 4� [Na2(H�2L4)]: To a stirred solution of 4 (45 mg,0.1 mmol) in dry MeOH (2 mL), was added NaOH (8 mg,0.2 mmol), previously dissolved in MeOH (1 mL). The mixture wasstirred at room temperature for 12 h. The solvent was then evapo-rated and the residue dried under reduced pressure for 48 h to give4� as a white solid (48 mg, 98%); m.p. 208�210 °C. MS (ESI posi-tive mode, H2O/MeOH): m/z (%) � 491.2 (10) [M � H]�, 469.3(76) [M � 2 H � Na]�, 447.3 (100) [M � 3 H � 2 Na]�.C22H40N10Na2 (490.6): calcd. C 53.86, H 8.21, N 28.54; found C53.87, H 8.16, N 28.57.

Syntheses of [Zn2(H�2L1)](ClO4)2 {[8](ClO4)2}, [Zn2(H�2L2)]-(ClO4)2·4H2O {[9](ClO4)2·4H2O} and [Zn2(L3)](ClO4)4·6H2O{[10](ClO4)4·6H2O} from Ligands L1�L3 (1�3). General Procedure:Working under argon, a solution of Zn(ClO4)2·6H2O (74 mg,0.2 mmol) in dry MeOH (1 mL) was added with stirring to a solu-tion of the polyamine ligand L1�L3 (0.1 mmol), dissolved in 2 mLof the same solvent. After 12 h, the precipitate was collected byfiltration and vacuum-dried at 100 °C over phosphorus pentoxidefor 24 h. In all cases, the white solid thus obtained was crystallisedfrom water.

{[8](ClO4)2}: Yield 55 mg, 77%. M.p. � 350 °C. MS (ESI positivemode, H2O): m/z (%) � 615.1 (10) [M � ClO4]�, 453.4 (100) [M� H � Zn(ClO4)2]�, 391.4 (75) [M � 3 H � 2 ZnClO4]�. MS(FAB�, 3-nitrobenzyl alcohol matrix): m/z (%) � 616.9 (20) [(M �

2 H)� � ClO4], 615.4 (12) [M� � ClO4], 553.2 (100) [(M � 2 H)�

� ZnClO4], 453.2 (47) [(M � H)� � Zn(ClO4)2].C18H32N10Zn2(ClO4)2 (718.2): calcd. C 30.16, H 4.47, N 19.55;found C 30.39, H 4.77, N 19.73.

{[9](ClO4)2·4H2O}: Yield 78 mg, 80%. M.p. � 350 °C. MS (ESIpositive mode, MeOH/H2O): m/z (%) � 893.2 (2) [M � H]�, 793.2(73) [M � ClO4]�, 731.4 (58) [M � 2 H � ZnClO4]�, 631.4 (50)[M � H � ZnClO4 � ClO4]�, 569.5 (100) [M � 3 H � 2 ZnClO4]�.MS (FAB�, 3-nitrobenzyl alcohol matrix): m/z (%) � 893.1 (23)[M � H]�, 793.2 (63) [M� � ClO4], 695 (100) [(M � H)� � 2ClO4], 731.3 (56) [M � H � Zn(ClO4)]�, 631.3 (77) [M � H �

Zn(ClO4)2]�. C27H46N14Zn2(ClO4)2·4H2O (968.5): calcd. C 33.48,H 5.62, N 20.24; found C 33.26, H 5.40, N 20.46.

{[10](ClO4)2·6H2O}: Yield 91 mg, 75%. M.p. � 350 °C. MS (ESIpositive mode, MeOH/H2O): m/z (%) � 995.1 (5) [(M � ClO4]�,833.0 (31) [M � H � Zn(ClO4)2]�, 733.5 (54) [M � Zn � 3 ClO4]�,671.2 (100) [M � 2 H � 2 Zn � 3 ClO4]�, 571.6 (91) [M � H � 2Zn(ClO4)2]�. MS (FAB�, 3-nitrobenzyl alcohol matrix): m/z (%) �

996.1 (20) [(M � H)� � ClO4], 995.1 (44) [M� � ClO4], 897 (33)[(M � 2 H)� � 2 ClO4], 895 (21) [M� � 2 ClO4], 733.3 (23) [M�

� Zn(ClO4)2 � ClO4], 633.3 (30) [M� � Zn(ClO4)2 � HClO4].C32H46N10Zn2(ClO4)4·6H2O (1207.5): calcd. C 31.84, H 4.81, N11.61; found C 31.91, H 5.17, N 11.69.

CuII and ZnII Coordination Chemistry of Pyrazole-Containing Polyamine Receptors FULL PAPERSynthesis of MII Dinuclear Complexes from the Disodium Dipyrazol-ate Salt 4�

Preparation of [Cu2(H�2L4)](ClO4)2 {[11](ClO4)2}: To a stirredsolution of the sodium dipyrazolate salt 4� (53 mg, 0.1 mmol) inMeOH (2 mL) was added a solution of Cu(ClO4)2·6H2O (74 mg,0.2 mmol) in 2 mL of the same solvent dropwise under argon. Themixture was stirred at room temperature overnight and the desiredcomplex [Cu2(H�2L4)](ClO4)2 {[11](ClO4)2} separated as a blueprecipitate. This was isolated by filtration, dried under vacuum andcrystallised from a water/methanol mixture (2:3) (16 mg, 21%); m.p.� 350 °C. MS (ESI positive mode, H2O/MeOH): m/z (%) � 671.0(100) [M � ClO4]�, 571.2 (58) [M � ClO4 � HClO4]�, 508.1 (9)[M � H � Cu(ClO4)2]�. C22H40Cu2N10(ClO4)2 (770.61): calcd. C34.28, H 5.23, N 18.17; found C 34.05, H 4.98, N 17.96.

Preparation of ZnII Dinuclear Complex [Zn2(H�2L4)](ClO4)2

{[12](ClO4)2}: To a stirred solution of the sodium dipyrazolate saltof 4� (53 mg, 0.1 mmol) in MeOH (2 mL) was added a solution ofZn(ClO4)2·6H2O (74 mg, 0.2 mmol) in 2 mL of the same solventdropwise under argon. The mixture was stirred at room tempera-ture overnight and the title complex precipitated. It was then iso-lated by filtration, dried under vacuum and crystallised from awater/methanol mixture (2:3) to afford [Zn2(H�2L4)](ClO4)2

{[12](ClO4)2} as a white solid (27 mg, 35%); m.p. � 350 °C. MS(ESI positive mode, H2O/MeOH): m/z (%) � 671.2 (32) [M �

ClO4]�, 609.3 (27) [M � 2 H � Zn � ClO4]�, 575.2 (51) [M � H� 2 ClO4H]�, 509.4 (100) [M � H � Zn(ClO4)2]�, 447.4 (64) [M� 3 H � Zn(ClO4)2 � Zn]�. C22H40N10Zn2(ClO4)2 (774.3): calcd.C 34.13, H 5.21, N 18.09; found C 33.99, H 5.24, N 17.97.

Synthesis of CuII Dinuclear Complexes from Ligand 4

Preparation of CuII Dinuclear Complex [Cu2L4](ClO4)4·2H2O{[13](ClO4)4·2H2O} (Blue and Red Forms): Working under argon,a solution of Cu(ClO4)2·6H2O (74 mg, 0.2 mmol) in dry MeOH(2 mL) was added with stirring to a solution of the polyamine li-gand 4 (45 mg, 0.1 mmol) in 2 mL of same solvent. After 4 h, ablue precipitate had formed. This was filtered off and vacuum-driedat 100 °C over phosphorus pentoxide for 24 h to give a blue solid(72 mg, 71%); m.p. � 350 °C. MS (ESI negative mode, H2O/MeOH): m/z (%) � 1070.9 (35) [M � ClO4]�, 968.8 (6) [M � H]�,869.0 (100) [M � (HClO4 � H)]�. MS (ESI positive mode, H2O/MeOH): m/z (%) � 871.0 (10) [(M � ClO4]�, 771.0 (10) [(M �

(HClO4 � ClO4)]�, 710.2 (30) [M � Cu(ClO4)2 � H]�, 671.1(100) [M � (2 HClO4 � ClO4)]�, 608.1 (5) [M � {Cu(ClO4)2 �

ClO4}]�, 547.3 (51) [M � {Cu(ClO4)2 � CuClO4} � 2 H]�.C22H40Cu2N10(ClO4)2(HClO4)2(H2O)2 (1007.56): calcd. C 26.23, H4.60, N 13.90; found C 26.21, H 4.54, N 13.69. When the aboveblue solid was heated to reflux in water (5 mL) for 5 min and theresultant blue aqueous solution allowed to stand at room tempera-ture, a red precipitate was unexpectedly obtained upon slow evap-oration of the solvent. This was filtered off, washed with methanoland vacuum-dried at 100 °C over phosphorus pentoxide for 24 h togive the complex {red-[13](ClO4)4·2H2O} as a brown-reddish solid(61 mg, 60%); m.p. � 350 °C. MS (ESI negative mode, H2O/MeOH): m/z (%) � 1071.4 (100) [M � ClO4]�, 969.4 (15) [M �

H]�, 869.4 (83) [M � (HClO4 � H)]�. MS (ESI positive mode,H2O/MeOH): m/z (%) � 871.0 (3) [M � ClO4]�, 771.0 (4) [M �

(HClO4 � ClO4)]�, 710.1 (22) [M � Cu(ClO4)2 � H]�, 671.1 (59)[M � (2 HClO4 � ClO4)]�, 608.1 (23) [M � {Cu(ClO4)2 �

ClO4}]�, 547.3 (33) [M � {Cu(ClO4)2 � CuClO4} � 2 H]�, 447.4(41) [M � 2 Cu(ClO4)2 � H]�, 224.3 (100) [{M � 2 Cu(ClO4)2}/2� H]�. MS (FAB�, 3-nitrobenzyl alcohol matrix): m/z (%) � 871.1(3) [M � ClO4]�, 771.2 (12) [M � (HClO4 � ClO4)]�, 710.3 (7)


[M � Cu(ClO4)2 � H]�, 671.2 (100) [M � (2 HClO4 � ClO4)]�,608.3 (5) [M � {Cu(ClO4)2 � ClO4}]�, 659.2 (44) [M � {3 HClO4

� ClO4}]�, 507.3 (7) [M � {Cu(ClO4)2 � 2 HClO4}·�].C22H42Cu2N10(ClO4)4·2H2O (1007.56): calcd. C 26.23, H 4.60, N13.90; found C 26.31, H 4.77, N 13.86.

Preparation of the Green CuII Trinuclear Complex[Cu3(H�2L4)](ClO4)4·2MeOH {[14](ClO4)4·2MeOH}: Basic anion-exchange resin (DOWEX, 550A OH) was added in portions to astirred solution of [4]·6HCl (100 mg, 0.14 mmol) in H2O (20 mL)until a pH of 11�12 was reached. The resin was then filtered offand the solvent removed under reduced pressure. The resultantsolid was crystallised from toluene to give 4 as a white solid of m.p.154�156 °C. Then, working under argon, a solution ofCu(ClO4)2·6H2O (93 mg, 0.24 mmol) in dry MeOH (1 mL) was ad-ded with stirring to a solution of the above polyamine L4 (4)(56 mg, 0.12 mmol) in 1 mL of the same solvent. After 4 h, a greenprecipitate was formed. This was filtered off and vacuum-dried at100 °C over phosphorus pentoxide for 24 h to afford compound[14](ClO4)4·2MeOH as a green solid; m.p. � 350. MS (FAB�, 3-nitrobenzyl alcohol matrix): m/z (%) � 927.7 (6) [M � (2 MeOH� ClO4H � 4 H)]�, 867.7 (14) [M � (2 Me � 2 ClO4)]�, 827.7(80) [M � (2 MeOH � 2 ClO4H � 4 H)]�, 732.1 (59) [M �

(2 MeOH � 2 ClO4H � ClO4)]�, 671.3 (51) [M � {2 MeOH �

Cu(ClO4)2 � ClO4}]�, 633.2 (100) [M � (2 MeOH �2 ClO4H �

ClO4)]�, 569.3 (37) [M � {2 MeOH � Cu(ClO4)2 � HClO4H �

ClO4}]�. MS (MALDI-TOF, 2,5-dihydroxybenzoic acid matrix):m/z (%) � 851.2 (16) [(M � (Me � MeO � 2 ClO4)]�, 797.2 (27)[M � 3 ClO4 � ClO4H]�, 750.3 (7) [M � (MeOH � Me � 3ClO4)]�, 734.3 (78) [M � (2 MeOH � 3 ClO4)]�, 672.3 (69) [M �

{2 MeOH � Cu(ClO4)2 � ClO4}]�, 649.4 (26) [M � (MeOH �

Me � 4 ClO4)]�, 635.4 (100) [M � 2 MeOH � 4 ClO4]�.C22H40Cu3N10(ClO4)4·2MeOH (1097.14): calcd. C 26.27, H 4.41,N 12.77; found C 26.06, H 4.08, N 12.38.

X-ray Crystallography: Crystal data and details of the data collec-tion for the complexes [Zn2(H�2L1)](ClO4)2 {[8](ClO4)2},[Cu2(H�2L4)](ClO4)2 {[11](ClO4)2} and [Zn2(H�2L4)](ClO4)2

{[12](ClO4)2} are given in Table 13. The crystal data were collectedwith a Nonius Kappa CCD instrument (λ � 0.71073 A). In the caseof [11](ClO4)2 and [12](ClO4)2 the data were collected at 120 K.The crystals were monitored for decay during data collection andappropriate corrections were made where necessary. A profileanalysis was performed on every reflection.[37] A semiempirical ab-sorption correction, Ψ-scan based, was performed in each case.[38]

Lorentz and polarisation corrections were applied and the datawere reduced to Fo

2 values. The structures were solved by Pattersonmethods using SHELXS-86 with an IBM PENTIUM II 300 com-puter.[39] Isotropic least-squares refinements were performed usingSHELXL-93.[40] The hydrogen atoms were placed in calculated po-sitions. All non-hydrogen atoms were refined anisotropically. Thehydrogen atoms were refined with a common thermal parameter.Atomic scattering factors were taken from the International Tablesfor X-ray Crystallography.[41] The molecular plots were producedusing ORTEP.[42] CCDC-240640, -240641 and -240642 contain thesupplementary crystallographic data for structures [8](ClO4)2,[11](ClO4)2 and [12](ClO4)2, respectively. These data can be ob-tained free of charge via www.ccdc.cam.ac.uk/const/retrieving.html[or from the Cambridge Crystallographic Data Centre, 12 UnionRoad, Cambridge CB2 1EZ, UK; Fax: (internat.) � 44-1223-336-033; or E-mail: [email protected]].

Electromotive Force (emf) Measurements: The potentiometric ti-trations were carried out in 0.15 m NaCl at 298.1 � 0.1 K usingthe experimental procedure (burette, potentiometer, cell, stirrer,


Table 13. Crystallographic data for [8](ClO4)2, [11](ClO4)2 and [12](ClO4)2

[8](ClO4)2 [11](ClO4)2 [12](ClO4)2

Empirical formula C19H26N6O8CL2Zn2 C21H36N10O8CL2Cu2 C21H36N10O8CL2Zn2

M 668.1706 754.5724 758.2984T [K] 293 293 293Crystal system triclinic orthorhombic orthorhombicColour colourless blue colourlessSpace group P1 Pmcn Pmcna [A] 10.4570(2) 13.762(1) 13.5330(7)b [A] 11.7980(2) 15.393(1) 15.3960(8)c [A] 13.3670(4) 15.524(2) 16.098(1)α [°] 111.624(1) 90 90β [°] 107.061(1) 90 90γ [°] 100.304(2) 90 90V [A3] 1387.06(5) 3288.6(5) 3354.1(3)Z 2 4 4Dcalcd. [g·cm�3] 1.60 1.524 1.502λ [A] 0.7103 0.7103 0.7103µ [mm�1] 1.974 1.513 1.645F(000) 680 1552 1560θ range of data collection [°] 1.83 � θ � 27.39 1.86 � θ � 21.76 1.83 � θ � 27.39Index ranges �13 � h � 12 �13 � h � 13 �17 � h � 17

�15 � k � 14 �16 � k � 16 �19 � k � 180 � l � 17 �16 � l �16 �19 � l � 20

Number of reflections measured 6230 13525 5893Unique reflections 630 2031 3348Number of reflections observed [I � 2σ(I)] 3995 1133 1424Parameters/restraints 393/0 245/0 261/0Goodness of fit/Goof 1.051 0.944 1.049R factor all reflections ωR2 0.135 0.151 0.2087R factor ωR1 0.055 0.078 0.080Max./min. nonassigned density [e/A3] 1.645/�0.742 0.919/�0.678 1.76/�1.49Weighting details: ω � 1/[σ2(Fo

2) � (0.1000P)2 � 0.0000P] where P � (Fo2 � 2Fc

2)1/3

microcomputer, etc.) fully described elsewhere.[43] The acquisitionof the emf data was performed with the computer program PA-SAT.[44] The reference electrode was an Ag/AgCl electrode in asaturated KCl solution. The glass electrode was calibrated as a hy-drogen ion concentration probe by titration of known amounts ofHCl with CO2-free NaOH solutions[45] and determination of theequivalent point by Gran’s method[46] which gives the standard po-tential E°� and the ionic product of water [pKw � 13.73(1)]. Thecomputer program HYPERQUAD[47] was used to calculate theprotonation and stability constants and the DISPO[48] program wasused to obtain the distribution diagrams. The titration curves foreach system (ca. 200 experimental points corresponding to at leastthree measurements, pH � 2�11, concentration of ligands 1 �

10�3 to 5 � 10�3 m) were treated either as a single set or as sepa-rated curves without significant variations in the values of the sta-bility constants. Finally, the sets of data were merged and treatedsimultaneously to give the final stability constants.

Magnetic Measurements: Magnetic susceptibility measurements ofpolycrystalline samples were measured over the temperature range2�300 K with a Quantum Design SQUID magnetometer using anapplied magnetic field of 1 T. Diamagnetic corrections of the con-stituents atoms were estimated from Pascal’s constants and a valueof 60 � 10�6 cm3·mol�1 was used for the temperature-independentparamagnetism of each copper(ii) ion.

Paramagnetic 1H NMR Spectroscopy: Paramagnetic 1H NMRspectra were recorded with a Varian Unity 400 spectrometer op-erating at 400 MHz. 1D spectra were recorded in (CD3)2SO withpresaturation of the (CH3)2SO signal during part of the relaxationdelay. Because the relaxation times were diverse, a variety of NMR


parameters were utilised. Acquisition times from 40 to 80 ms, relax-ation delay times from 50 to 200 ms, presaturation times from 50to 200 ms, spectral widths of 40 to 100 kHz and 256 to 1024 ofscans were used. 1D spectra were processed using exponential line-broadening weighting functions as apodisation with values of20�40 Hz. Chemical shifts were referenced to the residual solventprotons of (CD3)2SO resonating at δ � 2.50 ppm relative to SiMe4.1H longitudinal relaxation times were calculated using the inversionrecovery pulse sequence[49] (d1 � 180° � τ � 90° � acq), 18 valuesof τ were selected, (d1 � acq) values were 200 ms and the numberof scans was 512. The T1 values were calculated from the inversion-recovery equation. Transversal relaxation times were obtained bymeasuring the line broadening of the isotropically shifted signalsat half-height with the equation T2

�1 � π ∆ν1/2.

Molecular Modelling: Molecular modelling studies were carried outusing the AMBER method implemented in the Hyperchem 6.0package[50] and modified by the inclusion of appropriate param-eters. Where available, the parameters were analogous to those re-ported in the literature.[29,50] All others, namely those regarding theparticular atomic connections of our ligands with the metal ion,were developed according to the Norrby procedures[28,51] in a simi-lar way to the ones used in the MOMEC program.[52] The equilib-rium bond lengths and angles were derived from experimental val-ues on reasonable reference compounds. All additional parametersare given in the Supporting Information (Table S1). The startingstructures were built using Hyperchem capabilities and geometrieswere minimised to a maximum energy gradient of 0.1 kcal/(A mol)using the Polak�Ribiere (conjugate gradient) algorithm.

CuII and ZnII Coordination Chemistry of Pyrazole-Containing Polyamine Receptors FULL PAPERSupporting Information: Table S1 includes nonstandard force fieldparameters used in the modelling. Figures S1 (A�C) show the 1HNMR spectra of L4 at pH values 10.5, 8 and 3.7. Figure S2 rep-resents χM vs. T for compounds red-[13](ClO4)4·2H2O and[11](ClO4)2.

AcknowledgmentsFinancial support by the Spanish ‘‘Comision Interministerial deCiencia y Tecnologıa’’ (CICYT, SAF-99-0063 and BQU2003-09215-CO3-01) and Generalitat Valenciana (CTIDIB/2002/244 andGrupos03/196) is gratefully acknowledged.

[1] [1a] W. Kaim, B. Schwederski, Bioinorganic Chemistry: Inor-ganic Elements in the Chemistry of Life, An Introduction andGuide, John Wiley & Sons, Chichester, 1995. [1b]R. H. Holm,E. I. Solomon, Chem. Rev. 1996, 96, 2237�3042. [1c] I. Bertini,H. B. Gray, S. J. Lippard, J. S. Valentine, Bioinorganic Chemis-try, University Science Books, Mill Valley, California, 1994.

[2] [2a] J. P. Collman, Acc. Chem. Res. 1977, 10, 265�272. [2b] G.Feher, J. P. Allen, M. Y. Okamura, D. C. Rees, Nature(London) 1989, 339, 111�116. [2c] N. Kitajima, Y. Moro-oka,J. Chem. Soc., Dalton Trans. 1993, 2665�2671. [2d] P. Woolley,Nature (London) 1975, 258, 677�682. [2e] S. H. Gellman, R.Petter, R. Breslow, J. Am. Chem. Soc. 1986, 108, 2388�2394.[2f] T. Koike, S. Kajitani, I. Nakamura, E. Kimura, M. Shiro,J. Am. Chem. Soc. 1995, 117, 1210�1219. [2g] R. Breslow, Acc.Chem. Res. 1995, 28, 146�153.

[3] [3a] R. H. Holm, E. I. Solomon, Chem. Rev. 1996, 96,2237�3042. [3b] B. A. Mackay, M. D. Fryzuk, Chem. Rev. 2004,104, 385�401. [3c] T. Nakamura, Y. Yammamoto, Chem. Rev.2004, 104, 2127�2198.

[4] [4a] B. Rosenberg, L. Van Camp, K. Trigas, Nature (London)1965, 205, 698�699. [4b] B. Rosenberg, L. Van Camp, J. E. Tro-sko, V. H. Mansour, Nature (London) 1969, 222, 385�386. [4c]

C. Orvig, M. J. Abrams, Chem. Rev. 1999, 99, 2201�2842.[5] [5a] D. Parker, Chem. Soc. Rev. 1990, 19, 271�291. [5b] M. J.

Abrahams, B. A. Murrer, Science 1993, 261, 725�730. [5c] S.Jurisson, D. Berning, W. Jia, D. Ma, Chem. Rev. 1993, 93,1137�1156. [5d] M. J. Heeg, S. S. Jurisson, Acc. Chem. Res.1999, 32, 1053�1060. [5e] S. Kobayashi, H. Uyama, S. Kimura,Chem. Rev. 2001, 101, 398�3818.

[6] X. Huang, M. P. Cuajungco, C. S. Atwood, M. A. Hartshorn,J. D. A. Tyndall, G. R. Hanson, K. C. Stokes, M. Leopold, G.Multhaup, L. E. Goldstein, R. C. Scarpa, A. J. Saunders, J.Lim, R. D. Moir, C. Glabe, E. F. Bowden, C. L. Masters, D.P. Fairlie, R. E. Tanzi, A. I. Bush, J. Biol. Chem. 1999, 274,37111�37116.

[7] J. Elguero, ‘‘Pyrazoles’’ in Comprehensive Heterocyclic Chemis-try II, A Review of the Literature 1982�1995 (Eds.: A. R. Ka-tritzky, C. V. Rees, E. F. V. Scriven), Elsevier Science Ltd., Ox-ford, 1996, vol. 3.

[8] R. Mukherjee, Coord. Chem. Rev. 2000, 203, 151�218.[9] V. J. Aran, M. Kumar, J. Molina, L. Lamarque, P. Navarro, E.

Garcıa-Espana, J. A. Ramirez, S. V. Luis, B. Escuder, J. Org.Chem. 1999, 64, 6135�6146.

[10] L. Lamarque, C. Miranda, P. Navarro, F. Escartı, E. Garcıa-Espana, J. Latorre, J. A. Ramirez, Chem. Commun. 2000,1337�1338.

[11] L. Lamarque, P. Navarro, C. Miranda, V. J. Aran, C. Ochoa,F. Escartı, E. Garcıa-Espana, J. Latorre, S. V. Luis, J. F. Mira-vet, J. Am. Chem. Soc. 2001, 123, 10560�10570.

[12] [12a] A. Andres, M. I. Burguete, E. Garcıa-Espana, S. V. Luis,J. F. Miravet, C. Soriano, J. Chem. Soc., Perkin Trans. 2 1993,749�755. [12b] A. Bianchi, B. Escuder, E. Garcıa-Espana, S. V.Luis, V. Marcelino, J. F. Miravet, J. A. Ramırez, J. Chem. Soc.,Perkin Trans. 2 1994, 1253�1259. [12c] A. Andres, C. Bazzical-


uppi, A. Bianchi, E. Garcıa-Espana, S. V. Luis, J. F. Miravet,J. A. Ramırez, J. Chem. Soc., Dalton Trans. 1994, 2995�3004.

[13] J. A. Aguilar, E. Garcıa-Espana, J. A. Guerrero, J. M. Llinares,J. A. Ramırez, C. Soriano, S. V. Luis, A. Bianchi, L. Ferrini, V.Fusi, J. Chem. Soc., Dalton Trans. 1996, 239�246.

[14] A. Bencini, A. Bianchi, E. Garcıa-Espana, M. Micheloni, J. A.Ramırez, Coord. Chem. Rev. 1999, 188, 97�156.

[15] L. Behle, M. Neuburger, M. Zehnder, T. A. Kaden, Helv. Chim.Acta 1995, 78, 693�702.

[16] H. Weller, L. Siegfried, M. Neuburger, M. Zehnder, T. A.Kaden, Helv. Chim. Acta 1997, 80, 2315�2328.

[17] M. Kumar, V. J. Aran, P. Navarro, A. Ramos-Gallardo, A.Vegas, Tetrahedron Lett. 1994, 35, 5723�5726.

[18] M. Kumar, V. J. Aran, P. Navarro, Tetrahedron Lett. 1993, 34,3159�3162.

[19] M. Kumar, V. J. Aran, P. Navarro, Tetrahedron Lett. 1995, 36,2161�2164.

[20] [20a] S. F. Ralph, M. M. Sheil, L. A. Hick, R. J. Geue, A. M.Sargeson, J. Chem. Soc., Dalton Trans. 1996, 4417�4424. [20b]

A. Matsumoto, T. Fukumoto, H. Adachi, H. Watarai, Anal.Chim. Acta 1999, 390, 193�199 and references contained ther-ein. [20c] N. Yoshida, K. Ichikawaa, M. Shirob, J. Chem. Soc.,Perkin Trans. 2 2000, 17�26.

[21] E. L. Muetterties, L. J. Guggenberger, J. Am. Chem. Soc. 1974,96, 1748�1756.

[22] [22a] J. Ackermann, F. Meyer, E. Kaifer, H. Pritzkow, Chem.Eur. J. 2002, 8, 247�258. [22b] F. Meyer, A. Jacobi, B. Nuber,P. Rutsch, L. Zsolnai, Inorg. Chem. 1998, 37, 1213�1218.

[23] R. Zenobi, R. Knochenmuss, Mass Spectrom. Rev. 1998, 17,337�366.

[24] [24a] E. Lehmann, R. Knochenmuss, R. Zenobi, Rapid Com-mun. Mass Spectrom. 1997, 11, 1483�1492. [24b] R. Zenobi,Chimia 1997, 51, 801�803.

[25] P. Siemsen, U. Gubler, C. Bosshard, P. Guenter, F. Diederich,Chem. Eur. J. 2001, 7, 1333�1341.

[26] [26a] P. Comba, T. W. Hambley, Molecular Modelling of Inor-ganic Compounds, VCH Publishers, New Cork, 1995. [26b] C. R.Landis, D. M. Root, T. Cleveland, Reviews in ComputationalChemistry (Eds.: K. B. Lipkowitz, D. B. Boyd), VCH Pub-lishers, New York, 1995, vol. 6, pp. 73�148. [26c] M. Zimmer,Chem. Rev. 1995, 95, 2629�2649.

[27] W. D. Cornell, P. Cieplak, C. I. Bayly, I. R. Gould, K. M. J.Merz, D. M. Ferguson, D. C. Spelmeyer, T. Fox, J. W. Caldwell,P. A. Kollman, J. Am. Chem. Soc. 1995, 117, 5179�5197.

[28] D. E. Reichert, P.-O. Norrby, M. J. Welch, Inorg. Chem. 2001,40, 5223�5230.

[29] P. Comba, R. Remenyi, Coord. Chem. Rev. 2003, 238�239,9�20.

[30] I. Bertini, C. Luchinat, ‘‘NMR of paramagnetic substances’’,in Coord. Chem. Rev. 1996, 150.

[31] N. N. Murthy, K. D. Karlin, I. Bertini, C. Luchinat, J. Am.Chem. Soc. 1997, 119, 2156�2162.

[32] [32a] F. Meyer, S. Beyreuther, K. Heinze, L. Zsolnai, Chem.Ber./Recl. 1997, 130, 605�614. [32b] F. Meyer, A. Jacobi, L.Zsolnai, Chem. Ber./Recl. 1997, 130, 1441�1448.

[33] [33a] T. Kamiusuki, H. Okawa, N. Matsumoto, S. Kida, J.Chem. Soc., Dalton Trans. 1990, 195�198. [33b] V. P. Hanot, T.D. Robert, J. Kolnaar, J. G. Haasnoot, J. Reedijk, H. Kooij-man, A. L. Spek, J. Chem. Soc., Dalton Trans. 1996,4275�4281. [33c] J. Pons, X. Lopez, J. Casabo, F. Teixidor, A.Caubet, J. Rius, C. Miravitlles, Inorg. Chim. Acta 1992, 195,61�66. [33d] R. Gupta, R. Hotchandani, R. Mukherjee, Poly-hedron 2000, 19, 1429�1435. [33e] A. M. Schuitema, M. En-gelen, I. A. Koval, S. Gorter, W. L. Driessen, J. Reedijk, Inorg.Chim. Acta 2001, 324, 57�64.

[34] P. Roman, C. Guzman-Miralles, A. Luque, J. I. Beitia, J. Cano,F. Lloret, M. Julve, S. Alvarez, Inorg. Chem. 1996, 35,3741�3751.

[35] D. D. Perrin, W. L. F. Armarego, D. R. Perrin, Purification ofLaboratory Chemicals, Pergamon Press, Oxford, 1980.

E. Garcıa-Espana, P. Navarro et al.FULL PAPER[36] A. K. Covington, M. Paabo, R. A. Robinson, R. G. Bates,

Anal. Chem. 1968, 40, 700�706.[37] [37a] M. S. Lehmann, F. K. Larsen, Acta Crystallogr., Sect. A

1978, 30, 580�584. [37b] D. F. Grant, E. J. Gabe, J. Appl. Crys-tallogr. 1978, 11, 114�120.

[38] A. C. T. Nort, D. C. Phillips, F. S. Mathews, Acta Crystallogr.,Sect. A 1968, 24, 351�359.

[39] G. M. Sheldrick, C. Kruger, R. Goddard, Editors, Crystallo-graphic Computing, Clarendon Press, Oxford, England, 1985,p. 175.

[40] G. M. Sheldrick, SHELXL-93: Program for Crystal StructureRefinement, Institut für Anorganische Chemie, University ofGöttingen, Germany, 1993.

[41] International Tables for X-ray Crystallography, The KynochPress: Birmingham, England, 1974, vol. IV.

[42] C. K. Johnson, ORTEP; Report ORNL-3794, Oak RidgeNational Laboratory, Oak Ridge, TN, 1971.

[43] E. Garcıa-Espana, M.-J. Ballester, F. Lloret, J. M. Moratal, J.Faus, A. Bianchi, J. Chem. Soc., Dalton Trans. 1988, 101�104.

[44] M. Fontanelli, M. Micheloni, Proc. 1st Spanish-Italian Congr.Thermodynamics of Metal Complexes, Penıscola, Castellon(Spain), June 3�6, 1990.


[45] M. Micheloni, A. Sabatini, A. Vacca, Inorg. Chim. Acta 1977,25, 41.

[46] [46a] G. Gran, Analyst (London) 1952, 77, 661�671. [46b] F. J.C. Rossoti, H. Rossoti, J. Chem. Educ. 1965, 42, 375�378.

[47] P. Gans, A. Sabatini, A. Vacca, Talanta 1996, 43, 1739�1753.[48] A. Vacca, FORTRAN program for the determination of species

distribution in multicomponent systems, Florence, 1977.[49] A. E. Derome, Modern NMR Techniques for Chemistry Re-

search, Pergamon Press, Oxford, 1987.[50] Hyperchem 6.0 (Hypercube Inc), Scientific Software, USA,

2000.[51] [51a] P. V. Bernhardt, P. Comba, Inorg. Chem. 1992, 31,

2638�2644. [51b] J. E. Bol, C. Buning, P. Comba, J. Reedijk, M.Ströhle, J. Comput. Chem. 1998, 19, 512�523. [51c] P.-O.Norrby, P. Brandt, Coord. Chem. Rev. 2001, 212, 79�109. [51d]

C. Miranda, F. Escartı, L. Lamarque, M. J. R. Yunta, P. Nav-arro, E. Garcıa-Espana, M. L. Jimeno, J. Am. Chem. Soc. 2004,126, 823�833.

[52] T. W. Hambley, G. Laur, N Okon, MOMEC 97, Heidelberg,Germany, 1997.

Received August 2, 2004Early View Article

Published Online November 18, 2004

cuii and znii coordination chemistry of pyrazole-containing polyamine receptors − influence of the...

Documents